Engaging Patients in Tobacco Cessation Resources in Dental Settings: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Engaging Patients in Tobacco Cessation Resources in Dental Settings (NCT03656874), a randomized na study. The primary outcome is Percentage of Smokers With Delivery of Brief Interventions and/or Referral to Treatment.

2. Clinical Question

3. Trial Design

This was a completed, randomized, parallel, two-arm trial of clinical decision support compared with usual care. The brief description characterizes the design as a stratified, group-randomized controlled trial conducted in two settings. The registry record identifies clinics as the unit of randomization, but does not report the number of clinics or the details of the stratification.

Enrollment was 56. The posted primary-outcome groups include 13.0 participants assigned to usual care and 43.0 assigned to clinical decision support. These are the group counts for the primary outcome, not a description of how many clinics were randomized. The registry record does not explain the difference in group sizes or give further details on participant flow.

In the clinical decision support arm, providers received personalized scripts based on patients’ reported smoking attributes. In usual care, practitioners reviewed tobacco clinical guidelines during the consent process. Masking was listed as none. The registry record does not report further procedures for masking patients, providers, or outcome assessors.

The primary outcome was a patient-reported composite of brief intervention delivery or quitline referral within a week of the index dental encounter. Two secondary outcomes concerned initial actions within a week and long-term actions within six months. The record does not describe the trial’s original sample-size justification or its detailed statistical plan.

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: Randomization occurred at the clinic level.107 patients were authorized at their dental clinic to be contacted for the study. 51 patients declined to be consented or were not able to be reached; 56 participants consented to participate in the study.

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 Percentage of Smokers With Delivery of Brief Interventions and/or Referral to Treatment, measured over within a week of the index dental encounter.

The primary endpoint is the number of participants who reported receiving at least one qualifying action from a dental provider within a week of the index encounter. Qualifying actions included discussion of a quit plan, a quit date, medications to help with quitting, or strategies for quitting. A referral counted if the provider gave information about contacting a tobacco quitline or arranged for the quitline to contact the patient.

This is a composite endpoint: a participant was counted if any one of the listed actions was reported, rather than needing to report every action. The outcome therefore captures whether a provider delivered at least one brief intervention or referral, not the number, duration, or quality of the actions. It also does not directly measure whether a patient later quit smoking.

A patient-reported endpoint can capture care from the patient’s perspective, while relying on participants to recall and report what occurred. The registry record does not report further details about how reports were collected or verified. The endpoint fits the trial’s stated focus on provider delivery of brief interventions and referrals, but its interpretation should remain limited to that recorded care process and its stated time window.

7. Results Posted in the Registry

• Clinical Decision Support: 83.7% (36/43) versus 53.8% (7/13) on control; risk difference 29.9 percentage points (95% CI 0.6 to 59.1), risk ratio 1.55 (0.92–2.62); unadjusted two-proportion z-test p = 0.025.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Percentage of Smokers With Initial Actions Related to Cessation

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Clinical Decision Support: 46.5% (20/43) versus 23.1% (3/13) on control; risk difference 23.4 percentage points (95% CI -3.9 to 50.8), risk ratio 2.02 (0.71–5.72); unadjusted two-proportion z-test p = 0.132.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Clinical Decision Support and the control. The interval gives the range of effects that remain compatible with the data.

Percentage of Smokers With Long-term Actions Related to Cessation

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Clinical Decision Support: 69.2% (18/26) versus 40.0% (4/10) on control; risk difference 29.2 percentage points (95% CI -5.9 to 64.4), risk ratio 1.73 (0.78–3.86); unadjusted two-proportion z-test p = 0.107.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Clinical Decision Support and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes address related stages of cessation-related care: initial actions within a week and longer-term actions within six months. For both, the observed proportion was higher in the clinical decision support group than in usual care, and both posted risk-difference estimates were positive. Their confidence intervals include zero, and their unadjusted p-values are 0.132 and 0.107. Thus, the secondary comparisons show a direction broadly consistent with the primary point estimate, but they are less conclusive and do not supply a separate confirmation of it.

The secondary endpoints are not identical to the primary composite. They have their own definitions and follow-up windows, and the long-term outcome has fewer participants represented in each arm than the primary outcome. The registry record does not explain the difference in counts or report how missing or unavailable follow-up data were handled. These differences matter when comparing results across endpoints: apparent consistency in direction does not make the populations, measurements, or precision equivalent.

Multiplicity also affects how much weight to give secondary results. When several outcomes are tested, unadjusted p-values do not account for the opportunity to observe a low p-value by chance across the set. The registry record does not report a prespecified hierarchy, multiplicity adjustment, or confirmatory status for these outcomes. The secondary results therefore serve as supporting context about the overall pattern, rather than stand-alone evidence that the intervention changed each distinct outcome.

In combination, the secondary findings do not contradict the direction of the primary estimate, but their uncertainty and the unreported analysis details counsel against treating them as decisive. They describe related patient-reported actions, not smoking cessation itself.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Percentage of Smokers With Initial Actions Related to Cessation
• Percentage of Smokers With Long-term Actions Related to Cessation

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is binary: each participant either met the composite definition or did not. The posted counts give the number meeting the endpoint and the number in each arm. A risk difference expresses the contrast between the two arm proportions on an absolute scale. Its estimate is 0.299 for clinical decision support compared with usual care. The registry record also posts a risk ratio and an odds ratio, which express relative contrasts. Each measure summarizes the observed association differently; the scale should be kept in mind when interpreting its estimate and confidence interval.

The posted p-value for the primary comparison is unadjusted. In general, a p-value describes how compatible the observed data are with a specified null hypothesis under the assumptions of the statistical procedure. It is not the probability that the null hypothesis is true, and it does not measure the size or practical importance of an effect. The registry record does not identify the test or model used to produce the comparison statistics, so the analysis method cannot be described more specifically.

In general, a confidence interval describes a range of estimates compatible with the data and the method’s assumptions. For a risk difference, zero represents no difference on that scale. The primary interval runs from 0.006 to 0.591, so its endpoints are positive. This interval is still broad, and the estimate is not precise enough to identify a narrow range for the magnitude of the contrast. The posted p-value and interval are related summaries of uncertainty, not separate guarantees about the result.

The randomization was at the clinic level, according to the registry description. General statistical principle: outcomes among participants treated in the same clinic may be more alike than outcomes among participants in different clinics. When that clustering exists, analyses may need to account for it because treating all participants as fully independent can understate uncertainty. The registry record does not report the analysis model, any adjustment for clustering, or the number of clinics. It is therefore not possible to determine from the record whether or how clustering was incorporated into the posted estimates.

There are no reported analyses listed in the registry record, although arm counts and comparison statistics are posted. No adjusted analyses, subgroup analyses, interim analyses, or sensitivity analyses are described in the record. Their methods and results are therefore not reported in the registry record. The registry record also does not report the specific test, covariate adjustment, or approach to missing outcomes. The unadjusted results should not be recast as adjusted estimates or as evidence that these analysis issues were resolved.

The secondary endpoints are also binary participant counts, but they refer to different definitions and time windows. Their risk differences are accompanied by confidence intervals that include zero, and their unadjusted p-values are 0.132 and 0.107. In general, when several outcomes are examined, the chance of at least one apparently notable result can rise unless the analysis plan accounts for multiple testing. The registry record does not report a multiplicity strategy or a hierarchy among outcomes, so the secondary comparisons are best treated as exploratory context rather than independent confirmatory evidence.

Planning figures are not observed trial results. In general, a sample-size calculation uses assumptions about the effect to detect, variability, significance level, allocation, and analysis method. The registry record states assumptions for its planning illustrations, but those are not the trial’s original design assumptions. It does not report the original sample-size justification. The planning figures should not be interpreted as observed power or as a retrospective measure of how likely the trial was to detect an effect.

11. Confidence Intervals and Precision

The primary risk-difference interval extends from 0.006 to 0.591 around an estimate of 0.299. Because the interval is wide, the data are compatible with a range of effect magnitudes, from a value close to zero to a much larger positive contrast. Its lower endpoint is above zero, but that does not remove uncertainty about the size of the association. The unadjusted p-value of 0.025 is another summary of the posted comparison, not a substitute for considering the interval.

The other posted scales give a similarly cautious picture of magnitude. The risk-ratio interval, from 0.924 to 2.616, includes one, while the odds-ratio interval, from 1.134 to 17.14, is particularly broad. In general, a relative-effect interval including one is compatible with no relative contrast under that scale and method. The wide range for the odds ratio signals limited precision in that measure.

These intervals are based on modest group counts, and the registry record does not report whether the analysis accounted for clinic-level grouping. The record also does not report interval-construction methods. Those omissions limit how fully the precision estimates can be assessed. The posted intervals describe statistical uncertainty under the analysis that produced them; they do not account for every uncertainty about study conduct, measurement, or generalizability.

12. 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 planning illustrations give participants per arm for 80% power and participants per arm for 90% power. They are not total event counts: the record specifies that only time-to-event planning figures would be total events, and these outcomes are not time-to-event outcomes. The illustrations use a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each row considers detecting the stated fraction of the observed effect. The trial’s own original design assumptions are not reported in the registry record.

The 80% power column describes the planned number of participants per arm under which the calculation would have 80% power to detect the stated effect, given its assumptions. The 90% column gives the corresponding number per arm for 90% power. Power is a property of a planned design under specified assumptions; it is not a measure of whether the observed result is true. These values are planning illustrations based on the observed effect, not observed power.

For each endpoint, the illustrations require more participants per arm when planning around a smaller fraction of the observed effect. This is a general feature of sample-size planning: smaller effects are harder to distinguish from random variation, so more participants are generally needed to detect them with the same planned power. The calculations do not show that the observed effect will recur in another trial, and they do not replace a prospective design justification based on assumptions chosen before outcomes are known.

13. 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 = 0.538; treatment proportion p₁ = 0.837; average p̄ = 0.688.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6553 = 1.2844.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6203 = 0.5221.
4. n = (1.2844 + 0.5221)² / (0.2987)² = 36.6, rounded up to 37 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 <- 0.538461538461538                          # control proportion
p1 <- 0.837209302325581                          # 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(3656874)
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 = 37 (80% power), 49 (90% power)
 75% of observed effect: n per arm = 70 (80% power), 94 (90% power)
 50% of observed effect: n per arm = 166 (80% power), 222 (90% power)
power.prop.test: 36.6 
Simulated power at n = 37 per arm: 0.820 (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.

14. Multiplicity, Missing Data and Analysis Populations

The registry record lists a primary outcome and two secondary outcomes, but does not report whether the analyses followed a prespecified hierarchy or whether any adjustment for multiple comparisons was planned. It also does not report which outcomes, if any, were intended to be confirmatory. Because the posted secondary p-values are unadjusted, they should be interpreted with awareness that examining several outcomes can increase the chance of a nominally notable result.

Missing-data handling is not reported in the registry record. The primary outcome uses 13.0 participants in usual care and 43.0 in clinical decision support. The secondary long-term outcome uses 10.0 and 26.0 participants, respectively. The record does not explain why these counts differ, whether participants lacked follow-up, how missing reports were treated, or whether any imputation or sensitivity analysis was used. The count difference alone does not establish the reason or nature of missingness.

The analysis populations are also not described beyond the posted arm counts. The registry record does not state whether analyses followed an intention-to-treat approach, how participants were assigned to analysis populations, or whether exclusions occurred. It does not report the clinic-level analysis model or a method to account for group randomization. These details are particularly relevant because the trial is described as clinic-randomized.

The registry record contains no listed reported analyses, and does not describe interim analyses, covariate adjustment, subgroup analyses, or the trial’s full statistical analysis plan. These items should be treated as unreported rather than assumed absent from the original conduct. Their absence from the record limits assessment of whether the posted comparisons match prespecified methods and how robust they are to alternative assumptions.

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.

The adverse-event figures list zero serious adverse events and zero other adverse events in both groups, and zero deaths in each group. The listed at-risk counts are 13 for usual care and 43 for clinical decision support for both serious and other adverse events. These are the counts recorded in the registry entry.

These figures report what was entered for the listed groups; they do not establish that no adverse events occurred outside the recorded definitions or observation procedures. The registry record does not provide further detail about event collection, severity criteria, follow-up completeness, or safety analyses. The absence of listed events should therefore be read as a feature of the posted safety record, not as a broad conclusion about all possible harms.

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

The primary result describes a difference in patient-reported provider actions in this trial’s dental settings. The posted counts and risk-difference estimate favor more reports of at least one brief intervention or quitline referral in the clinical decision support arm. The confidence interval indicates uncertainty about the magnitude, and the comparison does not establish how the intervention would perform in other settings or populations.

The endpoint is a care-process measure. It does not measure smoking cessation, long-term health, or whether the patient used a quitline, followed a quit plan, or used medication. No such outcome is reported as the primary endpoint. Accordingly, the result should not be translated into a claim that the intervention changes smoking behavior or health outcomes.

The intervention was a clinical decision support approach using personalized scripts, while usual care involved review of tobacco guidelines during consent. The observed contrast applies to those trial conditions. It does not isolate which element of the intervention accounts for the difference, and the registry record does not report implementation measures that would show how consistently scripts were used.

Randomized allocation supports a comparative interpretation, but important design and analysis details remain unreported, including the number of clinics and whether the posted analysis accounted for clinic-level grouping. The posted unadjusted estimate should be understood within those limits. The appropriate conclusion is about the observed primary endpoint in the posted record, with its statistical uncertainty—not a general causal claim beyond that outcome and context.

17. Why This Trial Matters Statistically

This trial illustrates why statistical interpretation must begin with the outcome definition. A composite that counts any one of several provider actions answers a practical process question, but does not answer whether participants quit smoking or whether all components occurred equally often. The distinction matters when translating a favorable contrast into claims about what changed.

The design also illustrates the importance of matching analysis to randomization. Here, randomization is described at the clinic level, while the outcome summaries are participant counts. In general, participants grouped within clinics may have correlated outcomes. Without reporting the clinic count and analytic approach, readers cannot fully assess how the design relates to the posted uncertainty.

Finally, the primary estimate is more informative when read alongside its confidence interval, secondary outcomes, and missing analysis details. The primary interval is positive but broad; secondary comparisons have intervals including zero; and the registry does not report a multiplicity plan or missing-data method. These features do not erase the posted observations, but they affect the strength and scope of the conclusions that can reasonably be drawn from them.

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 registry record leaves several trial-specific details unclear. It does not report the number of randomized clinics, the stratification variables, or the method used to account for clinic-level grouping in the analysis. Because the randomization was at the clinic level, these details would help readers understand the statistical precision and the relation between the design and the posted estimates.

The outcome relies on patient reports of provider actions, and the record does not describe the reporting instrument, timing of collection, or validation procedures. It also does not explain why the long-term endpoint has smaller group counts than the primary endpoint or how unavailable follow-up was handled. These omissions limit assessment of measurement and missing-data risks.

The registry record does not report the original sample-size rationale, detailed statistical analysis plan, analysis populations, multiplicity strategy, or adjusted analyses. It is therefore not possible to determine from this record whether the posted analyses reflect prespecified procedures or whether alternative analyses were conducted. The posted primary comparison is unadjusted, and the record does not explain whether it incorporates the group-randomized design.

The trial took place in two settings, but the registry record does not provide details sufficient to judge how representative those settings are of other dental practices. The primary endpoint captures a narrow care process over a week; it does not directly assess cessation or longer-term health. These limitations constrain both the statistical interpretation and the scope of inference.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Engaging Patients in Tobacco Cessation Resources in Dental Settings — NCT03656874.
• Registry record last updated 2022-06-28; status last verified 2022-03.
• Results first posted 2022-06-28.
• 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

NCT03656874, “Engaging Patients in Tobacco Cessation Resources in Dental Settings,” is a completed randomized, parallel, two-arm, clinic-randomized trial comparing clinical decision support with usual care. Enrollment was 56, and masking was listed as none. The primary endpoint was patient-reported delivery of a brief intervention or quitline referral within a week of the index dental encounter.

The primary posted risk difference was 0.299, with a 95% confidence interval from 0.006 to 0.591 and an unadjusted p-value of 0.025. The secondary comparisons were directionally consistent but had confidence intervals including zero. The registry record does not report the original sample-size justification, detailed analysis plan, missing-data approach, or whether the primary estimate accounted for clinic-level grouping. The result concerns reported provider actions, not smoking cessation or health outcomes.