Smoking Cessation Intervention in Pregnant Women: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Smoking Cessation Intervention in Pregnant Women (NCT03024606), a randomized na study. The primary outcome is Exhaled Carbon Monoxide Levels.

2. Clinical Question

3. Trial Design

The registry identifies NCT03024606 as a randomized, open-label, parallel-group study of text messages focused on smoking cessation and pregnancy. The sponsor is St. Louis University. The phase is listed as NA. The recorded enrollment is 49, the status is completed, the start is listed as 2014-03, and primary completion is listed as 2017-06-14. The study title is “Smoking Cessation Intervention in Pregnant Women”; the official title is “Smoking Cessation Following Text Message Intervention in Pregnant Women.”

Participants were assigned to a texting group or a control group. The texting group received messages focused on smoking cessation and pregnancy; the control group received no text messages. Masking is listed as none. Randomization and an open-label assignment are therefore part of the registered design, but the registry record does not report details of sequence generation, concealment, or the randomization procedure.

The study summary describes an 18-month evaluation and says that all participants would have a standard-of-care smoking cessation visit. If the pharmacist, physician, and participant considered pharmacotherapy appropriate, nicotine replacement therapy patches or bupropion could be offered free of charge in 2-week intervals. The summary also describes bi-weekly visits for participants receiving pharmacotherapy, with additional vouchers, pharmacist meetings, and carbon monoxide testing until the medication course was complete. These are descriptions of the planned care; the registry record does not report actual use of pharmacotherapy, attendance, adherence to texts, or completion of visits.

The primary outcome is recorded at 1 month. Outcome counts are reported for 14 participants in the texting group and 16 in the control group, whereas the recorded enrollment is 49. The registry record does not explain this difference or provide a participant-flow account. It also does not report the trial’s original sample-size justification, its planned analysis populations, or a detailed statistical analysis 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: 23 patient were randomized to the texting group and 26 to the control. However, 9 in the texting group and 10 in the controls did not follow up for any study procedures of data collection, so we do not have data to report on them.

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 Exhaled Carbon Monoxide Levels, measured over 1 month.

The primary outcome is titled “Exhaled Carbon Monoxide Levels,” with the unit defined as participants and the measure defined as a count of participants with an event. The description clarifies that the outcome concerns self-reported cessation rates at 1 month, verified by exhaled carbon monoxide levels. The posted results give the number of participants meeting the outcome in each group: 8 of 14 in the texting group and 5 of 16 in the control group.

Smoking cessation is the outcome the intervention is intended to address. Self-report can capture a participant’s account of cessation, while verification using an exhaled carbon monoxide test adds a measured component to the outcome assessment. In general, using a verification measure alongside self-report can help distinguish a reported cessation outcome from one that also meets a measurement criterion. The registry record does not report the carbon monoxide threshold, the precise timing of testing within the 1-month timeframe, or how a self-report and test result were combined when they did not agree.

The registered outcome is a binary participant count rather than a continuous carbon monoxide measurement. The posted results therefore describe how many participants met the outcome, not the distribution of carbon monoxide readings. The outcome name and its description are not fully aligned in wording: the title names carbon monoxide levels, while the description specifies cessation rates verified by those levels. Further endpoint definition details are not reported in the registry record.

7. Results Posted in the Registry

• Texting Group: 57.1% (8/14) versus 31.2% (5/16) on control; risk difference 25.9 percentage points (95% CI -8.6 to 60.4), risk ratio 1.83 (0.78–4.31); unadjusted two-proportion z-test p = 0.153.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary result is a binary outcome: each participant is counted as having or not having the cessation outcome at 1 month. For a binary outcome, a group proportion is the number of participants meeting the outcome divided by the number with outcome data in that group. The registry reports 8 participants meeting the outcome in the texting group, with a group count of 14, and 5 in the control group, with a group count of 16. The reported proportions are 0.571 and 0.312, respectively. These figures describe the posted counts and do not indicate how missing outcomes were treated.

What does the risk difference express?

The risk difference subtracts the control-group proportion from the texting-group proportion. It is an absolute measure: it expresses the contrast in proportions on the probability scale. The reported estimate is 0.259. A positive value means the observed proportion was higher in the texting group under this direction of comparison; a negative value would mean it was lower. The 95% confidence interval, -0.086 to 0.604, communicates uncertainty around the estimate. Its breadth means the posted counts do not pin down the size of the difference precisely.

How do relative measures differ?

The risk ratio compares the two group proportions as a ratio, while the odds ratio compares their odds. The registry reports a risk ratio of 1.829 with a 95% confidence interval from 0.776 to 4.307, and an odds ratio of 2.933 with a 95% confidence interval from 0.657 to 13.09. Relative measures can appear numerically larger than an absolute difference because they use a different scale. They describe the same group data from different perspectives; the choice of measure should be clear, and an odds ratio should not be read as though it were a risk ratio.

What does the p-value say?

The unadjusted p-value is 0.153. In general, a p-value evaluates how compatible the observed data are with a specified null model, using the test’s assumptions. A larger p-value does not prove that groups are identical, and a smaller one would not by itself establish that an effect is important or unbiased. The registry record does not report the test method, a prespecified decision threshold for the primary analysis, or adjusted results. The stated two-sided significance level of 0.05 applies to the posted planning illustration, not necessarily to the original trial analysis.

Why do the counts matter?

When outcome counts are limited, estimates can be sensitive to the experience of individual participants, and confidence intervals can be broad. In general, sparse binary data may make estimates less stable and can limit the ability to distinguish among plausible effect sizes. The arm counts reported for the primary outcome are 14 and 16. The registry also records enrollment of 49 but does not explain how the outcome denominators relate to enrollment. Without participant flow or missing-data information, it is not possible to tell from the registry whether the outcome counts reflect nonmeasurement, exclusions, or another reason.

What does randomization contribute?

In general, random assignment is used to reduce systematic differences between groups at the start of a study, so that a comparison by assigned group can support causal interpretation when the design and analysis are appropriately conducted. Randomization does not guarantee that groups are identical, particularly in a small trial. The record identifies the study as randomized but does not describe the allocation sequence, concealment, or balance of baseline characteristics. Masking is listed as none. These missing details limit evaluation of the protections against bias.

Why can over-dispersion matter in other outcomes?

Over-dispersion is a general issue for count outcomes when observed variability exceeds what a chosen statistical model expects. It can make standard errors too small if the extra variability is ignored. The registered primary outcome is a participant count with a binary event, not a rate or repeated count outcome, and no dispersion estimate is reported. Over-dispersion is therefore not a reported feature of this result. More generally, the appropriate uncertainty method depends on the outcome scale, design, and analysis assumptions; those detailed methods are not reported in the registry record.

10. Confidence Intervals and Precision

The reported 95% confidence interval for the risk difference is -0.086 to 0.604, around a point estimate of 0.259. A confidence interval is a range of values compatible with the observed data under the statistical method and its assumptions. In general, the interval is not a guarantee that the true effect lies within its endpoints, nor does it assign a probability to the true effect after the data have been observed. Its usefulness is to show the uncertainty and range of effects that remain consistent with the analysis.

This interval includes values on both sides of zero. In general, zero is the value of no absolute difference for a risk-difference measure. The interval therefore does not isolate a single direction of difference. Its span also includes a wide range of possible magnitudes, so the point estimate should not be treated as a precise estimate of the effect.

The relative-effect intervals are also broad: the risk ratio is 1.829 with an interval from 0.776 to 4.307, and the odds ratio is 2.933 with an interval from 0.657 to 13.09. In general, the no-relative-difference value is 1 for a ratio measure. These intervals include values below and above 1, showing that the posted results leave substantial uncertainty about the relative contrast as well.

Precision is affected by the amount of outcome information and by variation in the data. The posted outcome group counts are 14 and 16, and the registry record does not explain their relationship to the recorded enrollment of 49. Without a participant-flow account or missing-data analysis, the precision estimates cannot resolve uncertainty about which enrolled participants contributed outcome information. The confidence intervals describe sampling uncertainty under the analysis that produced them; they do not account for possible bias from missing data, unmasked assessment, or other design limitations.

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 figures are planning illustrations based on the observed effect, not the trial’s original design justification. The registry record’s planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is specified as a stated fraction of the observed effect. The figures are participants per arm, not total events.

For planning at 100% of the observed effect, the illustration gives 57 participants per arm for 80% power and 76 participants per arm for 90% power. Planning for 75% of the observed effect gives 100 participants per arm for 80% power and 133 participants per arm for 90% power. Planning for 50% of the observed effect gives 219 participants per arm for 80% power and 293 participants per arm for 90% power.

In plain language, a power column describes the chance, under the planning assumptions and the effect size selected for planning, of detecting that effect using the specified statistical approach. The 90% column represents a higher planned chance of detection than the 80% column, and it therefore calls for more participants per arm at the same assumed effect. The rows illustrate that smaller effects need more participants to detect than larger effects, all else held to the stated assumptions.

These are planning figures, not observed power. They are calculated from the observed effect and do not establish what the original investigators intended or what the completed trial was designed to detect. The original trial’s sample-size justification and assumptions are not reported in the registry record. The figures should therefore be read as conditional illustrations, not as a reconstruction of the trial protocol.

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 = 0.312; treatment proportion p₁ = 0.571; average p̄ = 0.442.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.7023 = 1.3765.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6780 = 0.5707.
4. n = (1.3765 + 0.5707)² / (0.2589)² = 56.6, rounded up to 57 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 <- 0.3125                          # control proportion
p1 <- 0.571428571428571                          # 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(3024606)
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 = 57 (80% power), 76 (90% power)
 75% of observed effect: n per arm = 100 (80% power), 133 (90% power)
 50% of observed effect: n per arm = 219 (80% power), 293 (90% power)
power.prop.test: 56.6 
Simulated power at n = 57 per arm: 0.793 (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. Multiplicity, Missing Data and Analysis Populations

The registry record lists one primary outcome measure and no posted secondary outcomes. It reports zero posted analyses. The absence of a posted analysis entry means that analysis details such as the statistical test, model specification, covariate adjustment, and analysis population are not reported in the registry record. The unadjusted p-value and effect estimates are available, but they do not substitute for a full analysis plan.

Multiplicity arises when several outcomes, time points, treatment contrasts, or statistical tests are examined. In general, repeated testing can increase the chance of obtaining an apparently notable result by chance, and prespecified methods may be used to address that concern. The registry record does not report whether additional outcomes or analyses were planned or conducted, or whether any multiplicity adjustment was used. Because only one primary measure is posted, the extent of multiplicity cannot be assessed from the record alone.

The study description identifies 1 month as the primary outcome timeframe. It does not report whether other time points were assessed, whether the primary outcome definition changed, or whether any interim analyses occurred. Those details are not reported in the registry record and should not be inferred from the posted results.

The record also does not state whether the comparison followed an intention-to-treat approach, whether participants were analyzed according to assignment, or whether a per-protocol or other analysis population was used. It does not describe exclusions after randomization. The recorded enrollment is 49, while the outcome results use counts of 14 and 16; the registry record does not provide a flow diagram or explanation for the denominators.

Missing-data methods are not reported in the registry record. There is no account of how participants without a measured outcome were handled, whether missingness differed by group, or whether sensitivity analyses were conducted. In general, missing outcomes can affect both the estimated group contrast and its uncertainty, particularly when the likelihood of missingness is related to assignment or outcome status. Without these details, the posted estimates can be interpreted only for the counts shown, with important uncertainty about their relationship to the full randomized group.

14. 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 serious adverse event figures report zero affected participants in the texting group among 14 at risk and zero affected participants in the control group among 16 at risk. Deaths are reported as zero in each group. The figures for other adverse events also report zero affected participants in each group, with 14 at risk in the texting group and 16 in the control group.

These are counts in the registry’s posted safety results. They indicate that no participants are recorded as affected by a serious adverse event or other adverse event in these group summaries. They do not establish that no adverse events occurred outside the recorded results, and they do not show whether safety events were actively assessed in the same way for all participants. The registry record does not report the adverse-event collection period, definitions, assessment procedures, or detailed safety analysis methods.

The outcome denominators for safety are different from the recorded enrollment of 49, and the registry record does not explain this difference. Zero recorded events in these counts should therefore be read as a limited description of the posted safety information, not as proof of an absence of risk or as a comparison with enough information to characterize uncommon harms.

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

The posted results describe a one-month comparison of cessation outcomes in the randomized texting and control groups. The recorded outcome counts are 8 of 14 participants in the texting group and 5 of 16 in the control group. The estimated risk difference is 0.259, while its 95% confidence interval ranges from -0.086 to 0.604. Thus, the observed count-based contrast is positive, but its precision is limited and values in either direction remain compatible with the interval.

The result does not establish that the true group difference equals the point estimate. Nor does the unadjusted p-value of 0.153 establish that there is no difference. A p-value does not measure the probability that a treatment effect exists, and failure to obtain a small p-value is not proof of equivalence. The registry record reports no equivalence or non-inferiority objective, and the interval is not narrow enough to make a precise claim about the range of possible effects.

The study design compares assignment to text messages with assignment to no text messages, in a setting where the summary describes standard-of-care cessation visits for both groups and possible pharmacotherapy when considered appropriate. The registry does not report how many participants received medication, how closely participants followed the text intervention, or whether both groups received the other described care as planned. The observed contrast therefore cannot be separated into effects of particular care components from the registry results alone.

Randomized assignment supports a comparative design, but the record does not include allocation details, baseline characteristics, participant flow, or the analysis population. The outcome is described as self-reported cessation verified by exhaled carbon monoxide, but the verification criterion and detailed outcome handling are not reported. The open-label design is also relevant to interpreting a self-reported measure, although the record does not show whether awareness of group assignment altered reporting or measurement.

The results apply directly to the counts and outcome definition posted in the registry. They do not establish effects in all pregnant smokers, in groups not matching the described eligibility criteria, or at time points other than the stated 1 month. The recorded enrollment is 49, but the registry does not explain why outcome counts are reported for smaller group counts. The most defensible reading is that the available posted comparison is numerically higher in the texting group but remains uncertain, with important analysis and participant-flow details not reported.

16. Why This Trial Matters Statistically

This trial is statistically instructive because its primary result is a binary outcome with modest posted group counts and several common measures of comparative effect. The risk difference presents the absolute contrast, while the risk ratio and odds ratio present relative contrasts. Reading all three alongside their confidence intervals helps distinguish the scale of an estimate from the precision of that estimate. Here, each interval is broad, so no one measure removes the uncertainty evident in the posted counts.

The distinction between a point estimate and its uncertainty is central. The point estimate for the risk difference is positive, but the confidence interval crosses zero and covers a broad range. A single favorable-looking point estimate is not sufficient to convey what the data can support. Likewise, the unadjusted p-value should be interpreted with the effect estimate, interval, and study design rather than used as a binary judgment.

The registry also illustrates why a clear endpoint definition and participant accounting matter. The outcome title refers to exhaled carbon monoxide levels, while its description identifies self-reported cessation verified with carbon monoxide testing. The record does not state the test threshold or how the two components were combined. Enrollment is 49, but the outcome denominators are 14 and 16, with no explanation of the difference. Without these details, readers cannot fully assess endpoint classification or the contribution of missing outcomes.

Finally, the planning illustration makes explicit that sample-size requirements depend on the effect targeted and the desired power. It is distinct from a reconstruction of the investigators’ original design and from observed power. This distinction helps prevent a retrospective calculation from being mistaken for evidence that a completed trial was adequately sized for a particular clinical question. The registry record does not report the original sample-size justification or detailed analysis plan.

17. 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 limitations constrain interpretation. The registry describes an open-label randomized study, so masking was not used. The primary outcome description includes self-report, which may be affected by awareness of assignment in general. Carbon monoxide verification is also described, but the registry record does not report the threshold, testing procedures, or rules for reconciling self-report and test results.

The record gives an enrollment of 49 but reports outcome counts based on 14 participants in the texting group and 16 in the control group. It does not explain the participant flow, reasons for unavailable outcomes, exclusions, or analysis populations. This prevents assessment of how representative the posted outcome counts are of all randomized participants and how missing data may have affected the comparison.

The registry does not report baseline characteristics by group, allocation-concealment procedures, adherence to text messages, actual pharmacotherapy use, or attendance at the described follow-up visits. The summary describes standard care and possible nicotine replacement therapy patches or bupropion, but actual delivery and uptake are not reported. These factors limit the ability to understand what care participants received and how consistently the intervention was implemented.

The primary outcome is recorded at 1 month, and no secondary outcome results are posted. Longer-term cessation results are not reported in the registry record. The original sample-size justification, analysis plan, missing-data methods, and any interim analyses are also not reported. The combination of limited posted outcome counts, broad confidence intervals, and missing design and analysis details means that the effect estimate should be interpreted cautiously and as a description of the registry’s posted results rather than a complete account of the trial.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Smoking Cessation Intervention in Pregnant Women — NCT03024606.
• Registry record last updated 2018-10-24; status last verified 2018-10.
• Results first posted 2018-10-24.
• 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.

21. Record Summary

NCT03024606 is a completed, randomized, open-label, parallel-group study sponsored by St. Louis University. It compares text messages focused on smoking cessation and pregnancy with no text messages in pregnant women who smoked. The registry lists enrollment of 49, a start date of 2014-03, and primary completion on 2017-06-14. Phase is listed as NA.

The primary outcome is described as self-reported cessation at 1 month verified by exhaled carbon monoxide. The posted counts are 8 of 14 in the texting group and 5 of 16 in the control group. The risk difference is 0.259, with a 95% confidence interval from -0.086 to 0.604; the unadjusted p-value is 0.153. Serious adverse events, other adverse events, and deaths are recorded as zero in both group summaries.

The registry record does not report a posted analysis entry, a detailed analysis plan, missing-data handling, or an explanation of the difference between recorded enrollment and outcome denominators.