Evaluation of Oral Tobacco as a Harm Reduction Method for Smokers: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Evaluation of Oral Tobacco as a Harm Reduction Method for Smokers (NCT00711100), a non-randomized na study. The primary outcome is Product Preference.

2. Clinical Question

The study asked which oral product cigarette smokers would select, and described hypotheses involving taste, sensory characteristics, and nicotine content. It also aimed to examine product use when used as a switching tool, nicotine exposure, withdrawal symptoms, and potential for continued use. These are stated aims, not results: the posted record does not report findings for those questions.

The central statistical question is descriptive: how many participants selected each product? Preference counts can summarize choices in the study, but they do not by themselves establish why a product was chosen or whether a product changes smoking behavior. The crossover design means participants could encounter more than one product, while the registry does not give enough detail to reconstruct the sequence or compare product periods. The preference results are essential to answering the primary question, and they are not reported in the registry record.

3. Trial Design

The record describes a completed study with non-randomized allocation and a crossover model. It lists University of Minnesota as the sponsor, an enrollment of 104, and a status of completed. The recorded start was 2008-08 and primary completion was 2009-10. The phase is listed as NA. These descriptors identify the registered study but do not supply a full account of how it was conducted.

The intervention list includes Camel Snus, Marlboro Snus, Stonewall, Ariva, and General Snus. The descriptions identify Camel Snus, Marlboro Snus, and General Snus as oral smokeless tobacco products, and Stonewall and Ariva as oral dissolvable tobacco products. The registry also records nicotine amounts per portion for the products. The specific schedule, duration of exposure to each product, order of product use, and any washout period are not reported in the registry record.

A crossover design generally means that participants receive or experience multiple conditions over time, rather than being assigned to separate groups that each receive only one condition. That general feature can allow within-participant observations, but it also makes the sequence and timing of exposure important for interpretation. Here, the record does not report enough procedural detail to determine how such effects were handled.

Allocation is explicitly listed as non-randomized, and masking is listed as none. The record therefore does not describe a blinded, randomized comparison. It would be inappropriate to read product-specific counts as evidence from a like-for-like controlled comparison. The registry explicitly flags that the arms are not a like-for-like control comparison.

The enrollment figure and the safety tables’ at-risk counts are not the same. Reasons for this difference, participant flow, exclusions, and the number analyzed for the primary endpoint are not reported in the registry record. Nor are the protocol, sample-size justification, or detailed statistical plan reported. The study description gives aims and hypotheses, but those do not establish what analyses were ultimately performed.

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: Sampling Phase preceded the 2 week Abstinence Phase. Sampling phase involved sampling each study product for a 1/2 day; subjects then chose the product that they would like to use for a two week cigarette abstinence period

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 Product Preference, measured over 2 weeks.

The primary endpoint is titled Product Preference and is defined as the number of individuals who selected each product. Its stated timeframe is 2 weeks. The products named in the endpoint description are Camel Snus, Marlboro Snus, General Snus, Ariva, and Stonewall. The registry lists the endpoint unit as participants and its parameter as a number.

A count of selections is a direct descriptive measure of stated or recorded choice. It can show which products were selected in the study and how choices were distributed, if the counts are available. This endpoint was relevant to the study’s stated question about preferred products and its hypotheses involving taste, sensory aspects, and nicotine content. However, the registry does not report the observed counts, the method used to elicit or record preference, or whether participants could select more than one product.

Preference is not equivalent to use over time, successful substitution for cigarettes, reduced exposure, or changes in health. Those are distinct questions and would require their own outcomes and analyses. The record describes several secondary aims, but their results are not reported. Without the primary counts, the endpoint’s descriptive result cannot be assessed from the registry record. The posted results therefore do not permit a conclusion about which product was most often selected or how concentrated the choices were.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

This discussion explains the statistical ideas relevant to the registered design and distinguishes them from procedures that the registry does not describe. The record identifies a non-randomized crossover study, a participant-count primary endpoint, and safety event counts. It reports no analyses. The methods used to produce or analyze the outcomes are therefore not reported in the registry record.

What does a crossover design imply?

In general, a crossover design observes participants under more than one condition or product over time. Since one participant may contribute observations under multiple conditions, those observations are related rather than independent. A statistical analysis of crossover data often needs to account for this within-participant relationship. The order of exposure and the period in which an outcome is measured can also matter: responses may vary over time, and experience with one condition can potentially affect what is observed later. These are general considerations, not claims about what happened in this study. The registry does not report the product sequence, period structure, washout arrangements, or an analysis accounting for repeated observations.

Non-randomized allocation also limits what can be inferred from product comparisons. In general, random assignment helps balance participant characteristics across comparison groups, while non-random assignment can leave differences between groups or exposure periods. In a crossover study, the same participant may provide observations across products, but that does not remove the need to consider order, time, and other possible influences. Here, the record explicitly states that the arms are not a like-for-like control comparison. It does not report an adjustment strategy or a causal analysis.

What does a count endpoint tell us?

The registered primary endpoint is the number of individuals who selected each product. A count is a descriptive summary: it reports how many people meet a stated definition. To interpret it, a reader needs to know who was eligible to make a selection, how the selection was elicited, whether a participant could select more than one product, and how missing or unclear responses were handled. Those details are not reported in the registry record. The preference counts themselves are also not reported, so neither the observed distribution nor its uncertainty can be assessed.

Counts can be accompanied by statistical uncertainty when the goal is to infer beyond the observed participants. A confidence interval is a range of values generated by a statistical procedure; in general, it indicates the precision of an estimate under the assumptions of that procedure. It is not a guarantee that the true value lies in the interval. No confidence interval for the primary endpoint is reported in the registry record. Without the estimate and its uncertainty, readers cannot evaluate how precisely the study describes product preference.

How should event counts be understood?

The safety record gives, for each product, the number at risk and the number affected by serious and other adverse events. In general, an event count is a descriptive tally, not by itself a comparison of risk. Interpretation depends on the event definition, observation period, whether events were counted once per participant or repeatedly, and whether the same participants appear under multiple products. The registry gives a timeframe for the primary endpoint but does not report the safety observation period or detailed event-counting rules. The crossover design and repeated product listings mean the product rows should not be presumed to be independent groups.

In general, over-dispersion occurs when observed variability is greater than expected under a simple statistical model. It matters because a model that assumes too little variability can understate uncertainty and make results appear more precise than they are. Whether over-dispersion was relevant to any outcome in this study is not reported in the registry record. The record does not identify a model for the preference endpoint or event outcomes.

What can be said about planning?

Sample-size calculations are planning tools. In general, they describe how many participants might be needed under stated assumptions about an effect, variability, allocation, and an error threshold. They are not observed results and do not show that a study did or did not find an effect. Any sample-size illustrations elsewhere on the page should be read as planning scenarios, not as the study’s original design calculation or as observed power. The registry does not report a sample-size justification, the assumptions used by the investigators, or an interim-analysis plan.

The record’s stated analysis count is zero, and it does not report a statistical analysis plan or inferential results. This leaves important methodological questions unresolved: how repeated observations were handled, what estimates were targeted, whether uncertainty was quantified, and how missing observations were treated. General statistical principles can explain why those details matter, but cannot substitute for the missing study-specific methods.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("Camel Snus", "Marlboro Snus", "Stonewall", "Ariva", "General Snus")
n      <- c(98, 98, 98, 98, 98)
events <- c(27, 23, 24, 24, 0)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  Camel Snus  98  27  27.6  19.0  37.5  19.7  37.1  19.7  37.1
ARM  Marlboro Snus  98  23  23.5  15.5  33.1  16.2  32.8  16.2  32.8
ARM  Stonewall  98  24  24.5  16.4  34.2  17.0  33.9  17.1  33.9
ARM  Ariva  98  24  24.5  16.4  34.2  17.0  33.9  17.1  33.9
ARM  General Snus  98  0  0.0  0.0  3.7  0.0  3.8  0.0  3.7

11. Multiplicity, Missing Data and Analysis Populations

The study description includes a primary preference endpoint and several additional aims, including product use, nicotine exposure, withdrawal symptoms, continued use, and comparisons with participants’ own-brand cigarettes. Considering multiple outcomes or product comparisons can raise multiplicity concerns. In general, when many hypotheses are tested, the chance of at least one apparently unusual result can increase unless the analysis plan addresses that issue. The registry does not report a multiplicity strategy, a hierarchy of secondary outcomes, or which comparisons were performed. It also reports no analyses, so no adjustment approach can be identified.

Missing-data handling is likewise not reported in the registry record. The record lists enrollment of 104 and safety at-risk counts of 98 for each product, but does not explain the difference or state how many participants contributed to the primary preference endpoint. It does not describe withdrawals, incomplete product exposure, missing preference responses, or exclusions from analysis. These omissions prevent assessment of whether the analyzed participants differed from those enrolled.

The analysis populations are not described. The record does not state whether analyses included all enrolled participants, only those completing product exposure, or another defined group. For a crossover study, the handling of participants with incomplete sequences or unavailable measurements can affect which observations contribute to each product summary. The registry does not say how such cases were handled.

Because these analysis and reporting details are absent, the reader cannot determine whether the endpoint results were prespecified, how multiple outcomes were managed, or how missing observations may have influenced the findings. The absence of this information should not be interpreted as evidence that a particular procedure was or was not used.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 3% reporting threshold.

Most frequent non-serious adverse event terms

The registry safety summaries list 98 participants at risk for serious adverse events for each product, with zero affected in every listed product group. The reported figures therefore show no serious adverse events recorded in those summaries. Death counts are not reported in the registry record.

For other adverse events, the record lists 9 affected for Camel Snus, 5 for Marlboro Snus, 9 for Stonewall, 5 for Ariva, and 23 for General Snus, with 98 at risk for each. These are descriptive counts. The registry does not provide event definitions, severity details for these other events, observation periods, or information about whether individuals could contribute to counts for more than one product.

Because the study was non-randomized and crossover, the listed product groups should not be treated as independent randomized arms. The repeated at-risk count does not establish that different participants were observed for each product, nor does it establish equivalent exposure time. The counts do not by themselves support a conclusion that one product is safer than another. The registry’s safety figures should be read as reported event summaries, not as a controlled comparison or a complete account of risk.

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

The posted registry information supports a limited description of this study. It was a completed, non-randomized crossover study with enrollment of 104, and its primary endpoint was the number of individuals selecting each oral tobacco product over 2 weeks. The record identifies the products and reports adverse-event counts. It does not report the primary preference counts, so the central result cannot be summarized from this record.

This absence does not show that participants had no preferences, that products were equally preferred, or that the study found no difference. It means only that the preference results are not reported in the registry record. The same distinction applies to the additional aims. The study description includes questions about product switching, nicotine exposure, withdrawal, continued use, and dual use, but their results are not reported.

The safety summaries show zero serious adverse events among 98 at risk for each product, and report other affected counts by product. These data describe what the registry lists; they do not establish comparative safety. The crossover structure, non-randomized allocation, and lack of details about sequence and event ascertainment limit interpretation of between-product contrasts. Repeated product-level observations may involve the same participants, and the record does not explain how overlap was handled.

More broadly, descriptive counts do not establish cause and effect. They do not show that a particular product led to a particular choice or event, and they do not establish changes in smoking behavior or health. Those conclusions would require appropriate outcome measurements and analyses, details that are not reported in the registry record. Nor can the posted information establish statistical precision: no confidence intervals or inferential results for preference are reported.

The appropriate reading is therefore narrow. The record describes the design, endpoint, and some safety counts, while leaving the primary outcome and key methods unavailable. Readers should avoid turning missing results into a positive or negative finding. The results that would answer the registered preference question are not reported in the registry record.

14. Why This Trial Matters Statistically

This trial is statistically notable because the registered primary endpoint is straightforward to state—a count of participants selecting each product—yet the registry record does not report the counts needed to answer the primary question. This illustrates the difference between a clearly defined outcome and a usable result. Endpoint definitions matter, but interpretation also requires the observed data and enough information about how they were collected.

The crossover model adds a second statistical issue. In general, repeated observations from one participant are related, so treating them as if they came from separate independent groups can misrepresent uncertainty. Product order and measurement period may also matter. The registry does not describe the sequence or the analysis used to handle repeated observations. This makes clear why design labels alone cannot substitute for procedural and analytical detail.

The study is also explicitly non-randomized, and the record cautions against treating the arms as like-for-like controls. A table of product-level counts may look comparative, but the design and participant overlap determine what comparisons are justified. The safety figures are useful as descriptive reporting, but they do not establish differences in risk between products.

Finally, the record separates aims from evidence. Hypotheses about product preference, switching, nicotine exposure, and withdrawal are not results. Without reported estimates and measures of uncertainty, readers cannot assess the magnitude or precision of any preference pattern. This makes the study a useful example of why transparent outcome reporting, analysis methods, and participant accounting are necessary for statistical interpretation—not simply for presenting a study’s intended questions.

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 trial-specific limitations follow from the information available in the registry record. The primary preference counts are not reported, so the principal endpoint cannot be evaluated. The record gives no confidence interval, p-value, or other inferential result for that endpoint, and it reports no analyses. As a result, the magnitude and precision of any preference pattern are unknown.

The study is described as non-randomized and crossover, and the registry cautions that the arms are not a like-for-like control comparison. The sequence and timing of product exposure, any washout, and handling of repeated observations are not reported. These missing details limit interpretation of product-level comparisons. Masking is listed as none, but the record does not describe how participants’ choices or outcomes were elicited.

Participant accounting is incomplete in the registry record. Enrollment was 104, while the safety summaries list 98 at risk for each product. The reason for this difference and the primary analysis population are not reported. The record also does not describe missing outcomes, withdrawals, incomplete exposure, or how such information was handled.

The safety summaries lack reported event definitions and observation periods. It is unclear whether counts represent unique people or repeated events, and whether participants contributed to more than one product’s figures. Death counts are not reported in the registry record. These limitations mean the event numbers are descriptive and cannot support a product safety ranking.

Finally, the registry does not report the protocol, sample-size justification, analysis plan, or multiplicity approach. The study’s stated hypotheses and aims do not fill those gaps. This account therefore cannot establish how analyses were prespecified or whether all relevant results were reported.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Evaluation of Oral Tobacco as a Harm Reduction Method for Smokers — NCT00711100.
• Registry record last updated 2019-11-18; status last verified 2019-10.
• Results first posted 2015-09-16.
• 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.

19. Record Summary

The registry record describes a completed, non-randomized crossover study sponsored by University of Minnesota, with enrollment of 104. The recorded start was 2008-08, and primary completion was 2009-10. The phase is listed as NA, and masking as none.

The primary endpoint was Product Preference: the number of individuals selecting each product over 2 weeks. The interventions listed are Camel Snus, Marlboro Snus, Stonewall, Ariva, and General Snus. The preference counts and inferential analyses are not reported in the registry record.

Safety summaries list 98 at risk for each product, zero serious adverse events for each, and product-specific counts for other adverse events. The record does not explain the enrollment-to-safety-count difference or report deaths. The study description states aims involving preference and use, nicotine exposure, withdrawal, and continued use, but results for these aims are not reported.