This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Effect of Switching From Cigarette Smoking to the Use of IQOS on Periodontitis Treatment Outcome (NCT03364751), a randomized na study. The primary outcome is Periodontal Pocket Depth (PD) Change From Baseline at 6 Months.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT03364751 |
| Phase | NA |
| Condition | Periodontal Diseases, Chronic Periodontitis, Periodontal Pocket |
| Allocation | Randomized |
| Design model | Single Group |
| Masking | Single |
| Enrollment | 179 |
| Arms in the primary outcome | 3 |
| Lead sponsor | Philip Morris Products S.A. |
| Results posted | Yes |
2. Clinical Question
Population
Smokers with generalized chronic periodontitis receiving mechanical periodontal treatment. The primary endpoint description focuses on sites with an initial periodontal pocket depth of at least 4 mm. The analysis population was narrower than all randomized subjects: it included randomized subjects with product-use experience and a valid non-safety assessment, with some protocol deviations leading to exclusions.
Intervention
Switching from cigarette smoking to IQOS use. The registry describes IQOS as the Tobacco Heating System with Marlboro Heatsticks and notes that it is marketed in Japan under the IQOS brand.
Comparator
Continuing cigarette smoking. The study was described as randomized, controlled, parallel-group, and multicenter. The primary reported comparison is between the IQOS and Cigarette groups. The analysis also lists a Dual User group, reflecting the use-based analysis described in the record.
Primary question
At 6 months, what was the mean change from baseline in periodontal pocket depth at sites with an initial depth of at least 4 mm, and how did the reported result compare between the IQOS and Cigarette groups? The registry reports a mixed-model least-squares mean difference of 0.068, with a 95% confidence interval from -0.060 to 0.196 and p = 0.297.
The clinical question is about a measured periodontal outcome following mechanical treatment, not about every aspect of oral health. The registry’s brief summary states the purpose as comparing switching to IQOS with continuing cigarette smoking for response to periodontal therapy and overall oral health status. The posted primary endpoint is specifically the change in mean pocket depth at sites initially at least 4 mm deep. It therefore does not, by itself, represent all periodontal measurements, all oral-health outcomes, or longer-term outcomes.
Although the study description refers to two randomized groups, the primary analysis population is described according to actual self-reported product use. A Dual User group appears in the posted results, and a comparison between Cigarette and Dual User is also reported. These use-based groups should not be mistaken for additional randomized treatment arms: the registry does not explain in detail how participants came to be classified as dual users.
3. Trial Design
The registry describes a randomized, controlled, parallel-group, multicenter study with a 6-month timeframe. Its official title calls it open-label and specifies two arms: participants switching from cigarette smoking to IQOS and participants continuing cigarette smoking. The record lists enrollment as 179 and status as completed. The recorded study start was 2017-11-07, and primary completion was 2018-12-21. The phase is listed as NA.
The arm descriptions give approximately 86 patients in each assigned group. The primary endpoint results, however, are reported for IQOS, Cigarette, and Dual User groups, with analysis counts of 70, 84, and 17 respectively. These are not the same presentation as the two assigned arms in the study description. The analysis-population description says subjects were analyzed according to actual self-reported product use, and that some participants were excluded for protocol deviations. The registry does not fully explain the relationship between assigned arms and these analysis groups.
There is also a record-level classification inconsistency: the title describes a two-arm parallel-group study, while the model field is listed as single group. The masking field is listed as single, although the official title describes the study as open-label. The registry record does not reconcile these classifications, so they should be read as recorded metadata rather than as a complete account of conduct.
The sponsor is Philip Morris Products S.A. IQOS and Cigarette are both classified as interventions of type “Other.” The registry identifies the study population as smokers with generalized chronic periodontitis and describes the periodontal assessment as taking place following mechanical periodontal therapy. Details such as the specific treatment procedures, allocation concealment, and the operational conduct of follow-up are not reported in the registry record.
The posted primary comparison is IQOS versus Cigarette. A separate mixed-model comparison is reported for Cigarette versus Dual User. The registry does not establish that the latter comparison was randomized, and it should not be read as equivalent to the primary randomized contrast. The distinction matters because actual-use groups can differ from assigned groups, and comparisons between them may reflect factors beyond the planned intervention contrast.
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: 179 subjects were enrolled in the study. 172 subjects were randomized to the IQOS arm (n=87) or to the cigarette arm (n=85). The analysis was performed according to subjects' exposure (Product Use Category, i.e., IQOS; Cigarette; Dual use; or Other) over the study period as detailed in the "Arm/Group" (Reporting Groups) table.
Overall Study
| Stage | IQOS | Cigarette | Dual User | Other |
|---|---|---|---|---|
| STARTED | 70 | 84 | 17 | 1 |
| COMPLETED | 69 | 83 | 17 | 1 |
| NOT COMPLETED | 1 | 1 | 0 | 0 |
Reasons for not completing this period:
| Reason | IQOS | Cigarette | Dual User | Other |
|---|---|---|---|---|
| Withdrawal by Subject | 1 | 0 | 0 | 0 |
| Lost to Follow-up | 0 | 1 | 0 | 0 |
Analysis population for the primary outcome, as defined in the registry: The analysis population consisted of all randomized subjects with at least one product use experience and at least one valid non-safety assessment. Subjects were analyzed based on their actual self-reported product use. Some participants were excluded from analysis for protocol deviations (including, but not limited to, missing measurements). The "Other" reporting group, which included one subject, was not included in this analysis.
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.
Population: Subjects using during the course of this study over 30 HeatSticks, but less than 30 cigarettes monthly were classified as "IQOS" users. Monthly users of less than 30 HeatSticks, but over 30 cigarettes were classified as "Cigarette" users. Users of over 30 HeatSticks and over 30 cigarettes per month were classified as "Dual User". All other use patterns were classified as "Other".
| Characteristic | Unit | IQOS | Cigarette | Dual User | Other | Total |
|---|---|---|---|---|---|---|
| Participants analysed | 70 | 84 | 17 | 1 | 172 | |
| Age, Continuous | years | 48.1 (30 to 71) | 46.5 (30 to 77) | 54.4 (33 to 73) | 54.0 (54 to 54) | 48.0 (30 to 77) |
| Sex: Female, Male: Female | Participants | 13 | 17 | 3 | 0 | 33 |
| Sex: Female, Male: Male | Participants | 57 | 67 | 14 | 1 | 139 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 0 | 0 | 0 | 0 | 0 |
| Race (NIH/OMB): Asian | Participants | 70 | 84 | 17 | 1 | 172 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 0 | 0 | 0 | 0 | 0 |
| Race (NIH/OMB): Black or African American | Participants | 0 | 0 | 0 | 0 | 0 |
| Race (NIH/OMB): White | Participants | 0 | 0 | 0 | 0 | 0 |
| Race (NIH/OMB): More than one race | Participants | 0 | 0 | 0 | 0 | 0 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 0 | 0 | 0 | 0 | 0 |
| Region of Enrollment: Japan | participants | 70 | 84 | 17 | 1 | 172 |
| Nationality: Japanese | Participants | 70 | 84 | 17 | 1 | 172 |
| Nationality: Non-Japanese | Participants | 0 | 0 | 0 | 0 | 0 |
6. Primary Endpoint
The registry lists the primary outcome as Periodontal Pocket Depth (PD) Change From Baseline at 6 Months, measured over At 6 months.
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | 95% Confidence Interval |
| Unit | millimeters |
| Endpoint classification | Continuous Endpoint |
The primary endpoint is mean periodontal pocket-depth change from baseline at 6 months, measured in millimeters at sites with an initial pocket depth of at least 4 mm. Pocket depth is defined in the registry as the distance from the gingival margin to the point reached by a probe within the pocket. The outcome is continuous: each measurement can take a range of values, and the reported summary is a mean.
The endpoint is intended to describe change following mechanical periodontal therapy. A negative mean change indicates that measured pocket depth was lower at follow-up than at baseline. The posted mean changes are -1.046 for IQOS, -1.114 for Cigarette, and -1.177 for Dual User. These are group-level summaries; they do not describe the experience of every participant or every site.
Focusing on sites initially at least 4 mm deep defines the portion of the mouth being summarized and avoids treating all measured sites as interchangeable. The registry also lists other periodontal outcomes, including clinical attachment level and full-mouth measures. Those outcomes address related but distinct features of periodontal status and should not be substituted for the primary pocket-depth endpoint.
The primary endpoint is reported with 95% confidence intervals for each group and a mixed-model comparison for IQOS versus Cigarette. The registry does not report a threshold for a clinically important difference. Statistical interpretation can therefore describe the estimate and its uncertainty, but cannot determine from this record alone whether a particular difference would be clinically meaningful.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
For the primary endpoint, the posted mean change was -1.046 in the IQOS group and -1.114 in the Cigarette group. The group-specific 95% confidence intervals were -1.194 to -0.898 and -1.258 to -0.970, respectively. The registry’s mixed-model comparison of IQOS with Cigarette reports a least-squares mean difference of 0.068, a 95% confidence interval from -0.060 to 0.196, and p = 0.297.
The interval for the comparison includes zero. In statistical terms, the reported data do not distinguish the estimated contrast from zero at conventional significance thresholds. The interval also includes values on either side of zero, so it expresses uncertainty about the direction and size of the contrast. It is not evidence that the groups are identical, and it does not establish that any difference is absent. The p-value is a measure of how compatible the observed data are with a null model under the analysis assumptions; it is not the probability that the null hypothesis is true.
The estimate is a model-based least-squares mean difference, not simply a statement that one group’s raw mean is the other group’s mean plus a fixed amount. The record does not fully describe the model specification or the direction convention for the reported difference. Accordingly, the estimate should be quoted as reported rather than assigned a directional interpretation beyond the comparison as listed.
The Dual User group had a posted mean change of -1.177, with a 95% confidence interval from -1.399 to -0.956. The separate Cigarette-versus-Dual User mixed-model estimate is -0.064, with a 95% confidence interval from -0.273 to 0.146 and p = 0.550. This is a secondary comparison involving a use-based group, not the primary IQOS-versus-Cigarette contrast. The registry does not provide enough detail to treat it as a randomized comparison.
The outcome is measured in millimeters and pertains to a defined set of periodontal sites at a specific follow-up. The record does not provide a clinical-importance threshold, so statistical uncertainty cannot be translated into a judgment about clinical importance. Nor can these results establish what would happen beyond the recorded follow-up or for outcomes not measured by this endpoint.
Interpretation also depends on who contributed data. The primary population description requires product-use experience and at least one valid non-safety assessment, and says participants were analyzed according to actual self-reported product use. Some protocol deviations resulted in exclusions. These choices mean the posted estimate should not automatically be read as an estimate for every randomized participant. The registry does not describe missing-data handling sufficiently to assess how absent measurements affected the result.
9. Statistical Methods Explained
The primary endpoint is a continuous measure, and the registry reports group means, confidence intervals, and a model-based comparison. These summaries answer related but different questions: the means describe change within each reported group, while the comparison estimates the contrast between groups under the stated analysis model.
What does a confidence interval describe?
A confidence interval conveys uncertainty around an estimate. A narrower interval indicates greater precision under the model and analysis assumptions; a wider interval indicates less precision. The group-specific intervals for the primary endpoint describe uncertainty around each group mean change. The interval attached to the IQOS-versus-Cigarette estimate describes uncertainty around that between-group estimate. These intervals should not be confused with the range of individual participant outcomes: they concern estimates of group-level quantities.
The reported interval for the between-group comparison runs from -0.060 to 0.196, and the point estimate is 0.068. Because the interval includes zero, the data are compatible with a null contrast as well as contrasts in either direction within the interval. This does not prove equality. A confidence interval also does not mean that there is a 95% probability that the fixed, true parameter lies inside this particular calculated interval. The registry reports a 95% confidence level; the usual interpretation concerns the long-run performance of the interval procedure under repeated sampling and the model assumptions.
What is a least-squares mean difference?
A least-squares mean is a model-based group estimate. It can differ from a simple arithmetic comparison of posted group means because a model may account for specified features of the data. The registry identifies the method as mixed models analysis and the comparison parameter as a least-squares mean difference. It does not report the model formula, covariates, covariance structure, or all implementation details in the record. The reported estimate and interval can therefore be interpreted as the sponsor’s model-based comparison, but the record does not permit a full independent reconstruction of that model.
Mixed models are commonly used for data with repeated observations or clustered measurements, because observations from the same participant or other unit may be related rather than independent. Accounting for that dependence can improve estimation and uncertainty calculations when the model is appropriate. That is a general explanation of the method, not a claim about the exact structure used in this trial; the detailed model specification is not reported in the registry record.
Why distinguish within-group change from between-group comparison?
Each group’s mean change describes change relative to its own baseline. A between-group contrast asks whether the changes differ between groups. A change from baseline in one group, even when its confidence interval excludes zero, does not by itself answer whether the intervention group differs from the comparator group. The primary comparative inference should therefore focus on the reported IQOS-versus-Cigarette estimate and its uncertainty, alongside the group-specific summaries.
The posted group means are -1.046 for IQOS, -1.114 for Cigarette, and -1.177 for Dual User. These values describe the reported analysis groups; they do not replace the mixed-model comparison. No new contrast should be obtained by subtracting the posted group means and treating that result as the sponsor’s model estimate. The model-based estimate is reported separately.
What does the p-value add?
The primary comparison has p = 0.297. A p-value quantifies how unusual data at least as incompatible with a specified null model would be, assuming that model and the analysis assumptions hold. It does not measure the magnitude or clinical importance of a difference, and it is not the probability that the result occurred “by chance.” Here, the p-value is consistent with the confidence interval’s inclusion of zero. The estimate and interval are more informative about the range of contrasts compatible with the analysis than a threshold-only interpretation of the p-value.
The secondary Cigarette-versus-Dual User comparison has p = 0.550 and a confidence interval from -0.273 to 0.146. It has the same broad interpretive caution: the interval includes zero, and the comparison is based on actual-use groups as described in the record. It should not be presented as if it were a second randomized primary contrast.
What can planning illustrations tell readers?
Sample-size calculations are planning illustrations, not observed trial outcomes and not “observed power.” The registry’s stated planning assumptions use the observed effect, a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the target effect is expressed as a fraction of the observed effect. Such illustrations depend on those assumptions and can be useful for considering future study planning. They do not show that the trial was designed around those assumptions, and the registry record does not report the trial’s own sample-size justification. The illustration should not be interpreted as a retrospective measure of the probability that this completed study detected an effect.
How should the treatment-use groups be understood?
The study description presents two assigned groups, but the analysis population is described by actual self-reported product use. The results include a Dual User group, and the registry reports a Cigarette-versus-Dual User comparison. Reclassification by actual use can make the analysis population differ from the randomized groups. The record does not state the full rules for group classification, the timing of switching, or the handling of changes in use over time. Consequently, the model estimates should be interpreted in the context of the posted population definition rather than assumed to be a pure intention-to-treat comparison.
10. Analyses Reported by the Sponsor
| Method | Estimate | 95% CI | P-value |
|---|---|---|---|
| Mixed Models Analysis | LS Mean Difference 0.068 | -0.060 to 0.196 | 0.297 |
| Mixed Models Analysis | LS Mean Difference -0.064 | -0.273 to 0.146 | 0.550 |
The sponsor’s reported primary analysis is a mixed-model comparison of IQOS with Cigarette. It gives a least-squares mean difference of 0.068, a 95% confidence interval from -0.060 to 0.196, and p = 0.297. This model-based estimate is the principal comparative result posted in the registry.
The registry record does not contain a separate simple unadjusted recalculation for the primary two-group comparison. The recorded reason is that values were missing or not estimable for a two-arm comparison of this continuous endpoint. The group-specific means and confidence intervals remain available, but they cannot be treated as a substitute for the sponsor’s model-based contrast. In particular, no new arithmetic contrast should be presented as though it reproduced the mixed-model analysis.
The sponsor’s analysis and any unadjusted calculation would answer related but not necessarily identical questions: a model may account for features of the data that a direct comparison of summary means does not. Here, the detailed model specification is not reported in the registry record, limiting independent reproduction. The registry does not flag a discrepancy between a recalculation and the sponsor’s estimate. The sponsor’s reported result should therefore be described on its own terms, with its interval and p-value, and with the analysis-population and model-detail limitations kept in view.
11. Multiplicity, Missing Data and Analysis Populations
The record lists 18 outcome measures, 17 posted secondary outcomes, and 8 analyses. The primary endpoint is one part of a broader set that includes periodontal pocket depth, clinical attachment level, and full-mouth measures. When many outcomes or comparisons are examined, the chance of obtaining an apparently notable result by chance can increase. Whether and how that issue was addressed depends on the planned analysis strategy.
Multiplicity adjustment, hierarchy among endpoints, and control of an overall error rate are not reported in the registry record. The record also does not state whether the posted secondary analyses were prespecified, exploratory, or selected from a larger set of analyses. Secondary findings should therefore be interpreted as additional context rather than treated as independent confirmatory evidence without further information. The primary comparison should remain distinct from the other reported comparisons.
The primary analysis population is described as randomized subjects with at least one product-use experience and at least one valid non-safety assessment. Subjects were analyzed based on actual self-reported product use, and some participants were excluded for protocol deviations. The description of those exclusions is incomplete in the registry record; the available text ends after “including.” The specific deviations, the number of excluded participants, and their group distribution are not reported in the registry record.
Missing-data methods are also not reported in the registry record. It is not stated how missing baseline or follow-up measurements were handled, whether participants with incomplete outcome data contributed partially, or what assumptions were used for the mixed model. The posted counts and estimates cannot establish whether missingness was related to observed or unobserved outcomes. This limits assessment of the robustness of the reported results to incomplete follow-up.
The sample-size justification and any interim analyses are not reported in the registry record. The planning illustrations are based on stated assumptions and should not be treated as evidence about the trial’s original design decisions. The registry also does not report a multiplicity plan or a complete statistical analysis plan, so these aspects of the analysis cannot be independently evaluated from the record.
12. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| IQOS | 0 | 70 | 0.0% |
| Cigarette | 2 | 91 | 2.2% |
| Dual | 0 | 17 | 0.0% |
| Other | 0 | 1 | 0.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| IQOS | 0 of 70 |
| Cigarette | 0 of 91 |
| Dual | 0 of 17 |
| Other | 0 of 1 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| IQOS | 25 of 70 |
| Cigarette | 29 of 91 |
| Dual | 13 of 17 |
| Other | 1 of 1 |
Most frequent serious adverse event terms
| Event term | Organ system | IQOS | Cigarette | Dual | Other |
|---|---|---|---|---|---|
| Back pain | Musculoskeletal and connective tissue disorders | 0 of 70 | 1 of 91 | 0 of 17 | 0 of 1 |
| Jaw cyst | Musculoskeletal and connective tissue disorders | 0 of 70 | 1 of 91 | 0 of 17 | 0 of 1 |
Most frequent non-serious adverse event terms
| Event term | Organ system | IQOS | Cigarette | Dual | Other |
|---|---|---|---|---|---|
| Dental caries | Gastrointestinal disorders | 6 of 70 | 12 of 91 | 3 of 17 | 0 of 1 |
| Viral upper respiratory tract infection | Infections and infestations | 8 of 70 | 5 of 91 | 2 of 17 | 0 of 1 |
| Procedural Pain | Injury, poisoning and procedural complications | 1 of 70 | 4 of 91 | 1 of 17 | 0 of 1 |
| Gingival Pain | Gastrointestinal disorders | 2 of 70 | 3 of 91 | 1 of 17 | 0 of 1 |
| Toothache | Gastrointestinal disorders | 4 of 70 | 1 of 91 | 0 of 17 | 0 of 1 |
| Tooth fracture | Injury, poisoning and procedural complications | 2 of 70 | 1 of 91 | 1 of 17 | 1 of 1 |
| Hyperaesthesia | Nervous system disorders | 0 of 70 | 2 of 91 | 2 of 17 | 0 of 1 |
| Device failure | Product Issues | 1 of 70 | 1 of 91 | 1 of 17 | 0 of 1 |
| Cough | Respiratory, thoracic and mediastinal disorders | 0 of 70 | 0 of 91 | 2 of 17 | 0 of 1 |
| Gingivitis | Infections and infestations | 1 of 70 | 0 of 91 | 1 of 17 | 0 of 1 |
Serious events with the highest arm-level rate: back pain (1.1%), jaw cyst (1.1%).
The registry lists serious adverse events by product-use group. The serious-event affected counts are 0 for IQOS, 2 for Cigarette, 0 for Dual, and 0 for Other. The listed deaths are 0 in each of those groups. The record gives Back pain and Jaw cyst as the top serious-event terms, each with a reported value of 1.1%.
These figures describe the posted safety record, not a comparison that establishes the cause of an event. The record does not provide enough detail to assess event timing, duration, severity beyond the serious classification, or the basis for attributing events to product use. The displayed group counts also differ from the primary efficacy analysis counts: for example, the Cigarette serious-event group lists 91 at risk, while the primary Cigarette analysis lists 84. Safety and efficacy populations should not be assumed to be identical, and the registry does not fully reconcile these counts.
The record also lists other-than-serious adverse-event affected counts of 25 for IQOS, 29 for Cigarette, 13 for Dual, and 1 for Other. These are counts, not percentages, and should be read with their corresponding at-risk counts of 70, 91, 17, and 1. The short registry summary does not supply the detailed event-level context needed for a fuller comparative safety assessment.
13. What the Result Does, and Does Not, Mean
The posted result is an estimate of mean periodontal pocket-depth change at 6 months for sites that began at least 4 mm deep, in the analysis population described by the registry. The primary mixed-model comparison between IQOS and Cigarette is 0.068, with a 95% confidence interval from -0.060 to 0.196 and p = 0.297. The interval includes zero, so the analysis does not distinguish the reported contrast from zero under the model and its assumptions.
This finding does not establish that the groups are equivalent or that their true outcomes are identical. Demonstrating equivalence or non-inferiority would require a prespecified margin and a suitable analysis; those are not reported in the registry record. Nor does the result establish that either product improves or worsens health. It describes the posted endpoint and comparison, without making a treatment recommendation.
The result is specific to the selected periodontal sites, the recorded follow-up, and the analysis population. It does not by itself answer questions about clinical attachment level, all-mouth periodontal status, other oral-health outcomes, long-term effects, or outcomes in populations not represented by this study. The registry lists related secondary outcomes, but each must be considered on its own definition and analysis.
The actual-use analysis and inclusion criteria also matter. Participants were analyzed according to self-reported product use, and the record indicates that some were excluded for protocol deviations. The resulting groups may not correspond exactly to the randomized groups. Without fuller information on exclusions, missing data, and model specification, the estimate should be interpreted as the posted analysis rather than assumed to represent every randomized participant.
Finally, the registry does not report a clinical-importance threshold for the endpoint. Statistical uncertainty can be described using the confidence interval, but the record alone does not say what range of pocket-depth differences would be important to patients or clinicians. The study result is therefore best understood as a defined statistical comparison, with limits on both its precision and its scope.
14. Why This Trial Matters Statistically
This study raises a statistical question that is common in clinical research: how should a result be interpreted when the group-specific means appear close, but the comparative estimate remains uncertain? The primary contrast is the quantity that addresses the randomized treatment question, while the group means describe change within each reported group. Keeping those quantities separate prevents a within-group change from being mistaken for evidence of a between-group difference.
The record also illustrates why analysis populations matter. The planned description presents two randomized arms, but the primary results include a Dual User group and state that participants were analyzed according to actual self-reported product use. That distinction affects the interpretation of the estimates: an actual-use comparison is not automatically the same estimand as a comparison based on randomized assignment. Readers need the group definitions and analysis rules to understand what a reported contrast represents.
Precision is another central issue. The 95% confidence interval for the primary comparison spans values below and above zero. Rather than treating the p-value as a complete verdict, statistical interpretation should consider the estimate, its uncertainty, the endpoint scale, and the clinical context. The absence of a reported clinical-importance threshold leaves an important part of that context unspecified.
Finally, a registry record can document results while leaving key analytical details unavailable. Here, multiplicity procedures, missing-data methods, a complete statistical model specification, and the trial’s sample-size justification are not reported. Those omissions constrain independent appraisal and reproducibility. They do not erase the posted estimate, but they define how far readers can responsibly extend its interpretation.
15. 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 limitations are specific to the registry record for this trial. First, the study is described as randomized with two parallel arms, but the primary analysis is organized by actual self-reported use and includes a Dual User group. The record does not explain the full classification rules or reconcile the analysis groups with randomized assignment. This makes it difficult to determine precisely which treatment contrast the posted analysis estimates.
Second, the analysis-population description is incomplete. It states that randomized subjects needed product-use experience and a valid non-safety assessment, and that some participants were excluded for protocol deviations, but the description is truncated. The types and distribution of deviations and the number of exclusions are not reported in the registry record. The handling of missing outcome data is also not reported.
Third, the record does not provide the full mixed-model specification, including all model terms and assumptions. The posted least-squares mean difference can be reported, but readers cannot reproduce the analysis in detail from the registry alone. A separate unadjusted recalculation for the two-arm continuous comparison is listed as missing or not estimable, so there is no independent simple contrast to set alongside the sponsor’s model result.
Fourth, the registry lists 18 outcome measures, 17 posted secondary outcomes, and 8 analyses, but does not report a multiplicity plan. The possibility of selective emphasis among several outcomes and comparisons cannot be evaluated from the available record. The clinical importance of a particular pocket-depth difference is also not anchored by a reported threshold.
Finally, the record’s design classifications are not fully consistent: it describes a two-arm parallel-group study while listing a single-group model, and the official title calls the study open-label while the masking field is listed as single. The registry does not explain these entries. These limitations concern the completeness and interpretability of the record; they do not change the numerical results that are posted.
16. Related Tutorials
17. Related Statistical Calculators
18. Sources
- ClinicalTrials.gov: Effect of Switching From Cigarette Smoking to the Use of IQOS on Periodontitis Treatment Outcome — NCT03364751.
- Registry record last updated 2023-01-30; status last verified 2023-01.
- Results first posted 2021-02-18.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
19. Record Summary
NCT03364751 is a completed randomized study of switching from cigarette smoking to IQOS versus continuing cigarette smoking in smokers with generalized chronic periodontitis. The registry describes a 6-month, multicenter, parallel-group study and identifies periodontal pocket-depth change at 6 months as the primary endpoint.
For sites initially at least 4 mm deep, the posted mean changes were -1.046 for IQOS and -1.114 for Cigarette. The mixed-model IQOS-versus-Cigarette estimate was 0.068, with a 95% confidence interval from -0.060 to 0.196 and p = 0.297. The interval includes zero, and the registry does not report a clinical-importance threshold.
Interpretation is limited by the actual-use analysis population, the additional Dual User group, incomplete reporting of exclusions and missing-data handling, and absent detail on the statistical model and multiplicity plan. The record should be read as reporting a specific endpoint and analysis rather than a complete account of all design and analysis decisions.