Incorporating Veterans' Preferences Into Lung Cancer Screening Decisions: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Incorporating Veterans' Preferences Into Lung Cancer Screening Decisions (NCT02899754), a randomized na study. The primary outcome is Decisional Conflict.

2. Clinical Question

3. Trial Design

The registry identifies NCT02899754 as a completed, randomized, single-masked, parallel-group trial. The study title is “Incorporating Veterans' Preferences Into Lung Cancer Screening Decisions.” The sponsor was the VA Office of Research and Development. The listed enrollment was 140. The recorded start date was 2019-01-28, and primary completion was 2021-11-30.

The trial compared a behavioral intervention, the Lung Cancer Screening Decision Tool, with a behavioral control intervention. The control content concerned general disease prevention and health promotion unrelated to lung cancer, and was delivered through the same modality for a similar amount of time. The registry identifies the trial as randomized and single-masked but does not report who was masked or how masking was implemented. Further details of sequence generation and allocation concealment are not reported in the registry record.

Participants were veterans receiving primary care at a participating VA Medical Center who met the age and smoking-history eligibility criteria. The registry describes participants as being asked to link to a study website. The full details of recruitment, consent, and intervention completion are not reported in the registry record.

The primary endpoint was decisional conflict at one month following the intervention. Screening uptake at six and nine months appears among the posted secondary results. The registry also names knowledge, decision regret, anxiety, and lung cancer worry as secondary outcome areas. Results for all of those outcomes are not included in the posted figures discussed in this analysis. The protocol's planned analysis population, handling of missing data, and prespecified statistical analysis plan are not reported in the registry record.

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.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: There were 7 participants that were lost to follow-up at the 1 month visit in the experimental group and 9 participants that were lost to follow up at the 1 month visit in the control group.

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 Decisional Conflict, measured over 1 month following the intervention.

The primary endpoint was decisional conflict one month following the intervention, measured with the Decisional Conflict Scale. The registry describes a 16-item scale covering uncertainty, informed choice, value clarity, support, and effective decision-making. Items are scored and combined into a total scale value ranging from 0, indicating low decisional conflict, to 100, indicating high decisional conflict. Higher scale values therefore represent greater conflict, not better knowledge or more screening.

This measure is relevant to a screening decision because the decision may involve weighing possible benefits against harms and uncertainty. In general, a decisional-conflict measure can capture how uncertain or unsupported a person feels while considering a health decision. It is distinct from whether a person ultimately undergoes screening: a person may have a measured level of conflict without that score itself establishing which choice is appropriate.

The registry reports group means with 95% confidence intervals. The intervention group's mean was 25.7, with a 95% confidence interval from 21.4 to 30.1; the control group's mean was 29.9, with a 95% confidence interval from 25.6 to 34.2. These are arm-specific summaries, not a confidence interval for the difference between arms. The registry reports a two-sided t-test p-value of 0.18 for the comparison. The scale values and p-value describe the posted endpoint results; they do not by themselves establish the effect of the tool on decision quality or on subsequent screening.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry reports group means for the continuous primary endpoint and a two-sided t-test for the between-group comparison. In general, a mean summarizes the center of a set of measured values, while a t-test evaluates whether the observed group difference is compatible with a null comparison under the test's assumptions. The reported p-value of 0.18 is not the probability that the null hypothesis is true, nor is it the probability that the result occurred “by chance.” It describes how unusual data at least as incompatible with the null as the observed result would be under the test assumptions.

The primary endpoint is reported with 95% confidence intervals for each arm. In general, a confidence interval expresses uncertainty in an estimate under the method used to calculate it. The two intervals in the registry are intervals around separate group means, not an interval around the contrast between those means. The registry does not provide the standard error or confidence interval for the primary between-group effect. A reader should not infer that the difference is statistically significant or nonsignificant merely by checking whether arm-specific confidence intervals overlap; the direct comparison is the reported two-sided t-test.

For the binary uptake outcomes, the posted counts identify participants who underwent screening and the group denominators. A risk is the proportion with the event in a group. A risk difference compares the two group risks on an absolute scale. The recalculated risk differences are 0.166 at six months and 0.196 at nine months, with the confidence intervals reported alongside those estimates. These recalculations use the posted summary counts and are unadjusted: they do not account for any covariates or other features of the trial design. The risk-difference intervals describe the uncertainty in those unadjusted contrasts under the calculation method, not uncertainty in an adjusted model.

The registry also includes relative measures for the uptake results. A risk ratio compares the event risks as a ratio, whereas an odds ratio compares the odds of the event. These measures use different scales and can give a different numerical impression from an absolute risk difference. They should not be interpreted as interchangeable, and none of these unadjusted measures establishes causation beyond the design and analysis assumptions. The recalculated risk ratios and odds ratios are based on the posted counts, while the registry's reported analyses are separately recorded as a t-test at six months and a chi-squared test at nine months.

Why can reported and recalculated p-values differ?

A recalculated p-value is derived from the posted summary statistics using a specified simple, unadjusted comparison. The sponsor's result may reflect a different test or analysis procedure. At six months, the registry records a two-sided t-test with p = 0.18, while the unadjusted recalculation has p = 0.031. At nine months, the registry records a chi-squared analysis with p = 0.01, while the unadjusted recalculation has p = 0.015. The record does not provide enough detail to determine why the six-month result differs, or to establish the exact analysis implementation underlying each sponsor-reported value. A recalculation should therefore be understood as a check based on posted summaries, not a replacement for the trial analysis.

What do the planning illustrations mean?

The planning illustrations use the observed uptake effects as a starting point and consider detecting the full observed effect or a stated fraction of it. The stated assumptions are a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. Such calculations can illustrate how a target effect and design assumptions relate to the participants per arm for 80% or 90% power. They are not the trial's original sample-size justification, and the record does not report that these calculations were used to design this trial. They are not observed power: observed results cannot establish the probability that the completed trial detected an effect. The illustrations also do not account for every possible feature of a future study, such as missing outcomes or deviations from assumptions.

What assumptions remain unclear?

The registry record does not report a detailed statistical analysis plan, the analysis population for each endpoint, covariate adjustment, handling of missing values, or whether analyses were prespecified. It also does not describe any interim analyses. These omissions matter because analysis choices can affect estimates and uncertainty, especially for a trial with multiple outcomes and follow-up times. The posted results support descriptive interpretation of the reported endpoints, but they do not permit a full audit of the analysis from the registry record alone.

10. Analyses Reported by the Sponsor

For the primary endpoint, the registry reports a two-sided t-test with p = 0.18. It does not report a recalculated primary contrast or a confidence interval for that contrast in the posted results. The arm-level means and confidence intervals can be described directly, but they do not supply the missing treatment-effect interval. The sponsor's reported primary analysis is therefore the available inferential result for decisional conflict.

For uptake at six months, the registry reports a two-sided t-test with p = 0.18, while the unadjusted recalculation from the posted counts gives p = 0.031. At nine months, the registry reports a chi-squared result with p = 0.01, compared with the unadjusted recalculated p = 0.015. These values arise from distinct reported and recalculated analyses; details that would reconcile them are not reported in the registry record. Recalculated estimates are useful as transparent checks on the posted counts, but are not a substitute for a complete trial analysis.

The record does not flag a discrepancy between its results and recalculations. The registry's own analyses take precedence when interpreting the trial's reported inferential findings. The recalculated risk differences and their intervals remain useful descriptive, unadjusted summaries, provided they are not presented as adjusted or prespecified trial estimates.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists multiple outcome areas and reports analyses at more than one follow-up time. When several outcomes, time points, or comparisons are tested, the chance of obtaining at least one apparently notable result can increase. In general, multiplicity procedures aim to account for this family of comparisons, or a prespecified hierarchy can distinguish primary from secondary questions. The record does not report a multiplicity adjustment, a testing hierarchy beyond identifying decisional conflict as primary, or a statistical analysis plan explaining how secondary tests were to be interpreted.

The primary endpoint remains the clearest inferential focus because it is identified as primary and tied to the stated hypothesis about decisional conflict at one month. The secondary uptake results are relevant but should be interpreted with more caution, especially because the registry includes multiple secondary outcome areas and follow-up times. It is not reported whether the tests shown were prespecified, whether other analyses were conducted, or whether the reported results are a complete account of all analyses.

Missing follow-up is noted for the primary endpoint: 7 participants in the experimental group and 9 participants in the control group were lost to follow-up at the one-month visit. The registry reports 62 participants in each arm for the primary endpoint. It does not explain how missing primary outcome values were handled, whether any imputation was used, or whether the analysis was based on all randomized participants or only those with observed endpoint data. The reported endpoint population and the losses to follow-up should therefore be considered together, without assuming a particular missing-data method.

For uptake, the posted group denominators are 69 in the intervention group and 71 in the control group at both listed time points. The record does not explain how participants with unknown screening status, if any, were classified, or how missing follow-up was handled for these outcomes. The planned analysis populations and reasons for missingness are not reported in the registry record. Missingness can affect interpretation if the likelihood of an observed outcome differs by group or by participants' unobserved outcomes; whether that occurred cannot be determined from the posted information.

12. 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 registry reports zero serious adverse events affecting participants in the intervention group and zero in the control group. It also reports zero participants affected in the other adverse-event category for both groups, and zero deaths in both groups. The corresponding at-risk counts are 69 for the intervention group and 71 for the control group in each category. No top serious-event terms are listed.

These figures describe what is recorded in the registry's adverse-event results. They do not prove that no adverse experiences occurred, that every event was captured, or that the intervention is free of risk. The registry record does not provide further detail about event definitions, severity grading beyond the categories shown, collection procedures, or the period covered by these counts. The results should therefore be read as the posted event counts, not as a complete assessment of safety or as a comparison of all possible harms.

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

The primary result describes average decisional-conflict scores one month after the intervention. The intervention group's mean was lower than the control group's mean, but the registry-reported test has p = 0.18. That result does not provide strong statistical evidence of a group difference under the reported test. It also does not demonstrate equivalence, rule out a potentially meaningful difference, or establish that the decision tool had no effect. The registry does not report a confidence interval for the primary contrast or a threshold defining a meaningful change in the scale.

The uptake results describe screening behavior at later follow-up, not the quality of the decision or the health consequences of screening. The posted counts and unadjusted comparisons favor higher uptake in the intervention arm at both time points. They do not show whether participants understood the trade-offs, whether screening was appropriate for each individual, or whether uptake changed because of the tool rather than other factors. The record does not report outcomes that would establish the effect of the intervention on lung cancer diagnosis or mortality.

Because the study was randomized, its design can support a comparison between assigned groups, subject to the conduct and analysis of the trial. But the registry's limited information about missing data, analysis populations, and statistical methods restricts how completely those comparisons can be evaluated. The posted recalculations are unadjusted checks from summary counts; they are not substitutes for a full analysis plan or participant-level analysis.

In general, a non-small p-value is not proof of no effect, and a small p-value is not proof that an effect is important or unbiased. Interpretation depends on the estimate, its uncertainty, the outcome's meaning, the study design, and how many comparisons were considered. For this trial, the primary endpoint and secondary uptake outcomes answer different questions. Neither the primary finding nor the secondary results alone supports a broad claim about all aspects of informed screening decisions.

14. Why This Trial Matters Statistically

This trial is statistically informative because it places a decision-support intervention against an active information control and evaluates both a decision-related outcome and later screening behavior. Those outcomes need not move together. A tool may affect what participants know or how they deliberate without producing the same pattern in uptake, while uptake alone does not measure the quality of a decision. Keeping these constructs separate is important when interpreting behavioral interventions.

The primary endpoint is a continuous scale with a clear direction: higher decisional-conflict scores indicate greater conflict. The registry reports group means and arm-specific confidence intervals, along with a direct two-sided test. This creates an opportunity to distinguish the descriptive direction of the means from the strength of statistical evidence. It also illustrates why an estimate and its uncertainty are more informative than treating a p-value as a complete account of a result.

The uptake comparisons further show why the method attached to a result matters. Recalculations from posted counts yield unadjusted risk differences, while the registry lists its own tests. Those summaries are not identical analyses, and the record lacks detail needed to reconcile all differences. Transparent reporting of analysis populations, missing-data methods, prespecification, and multiplicity handling would make results easier to assess and reproduce.

Finally, planning illustrations based on observed effects can help readers understand the scale of future studies under stated assumptions, but they do not reveal whether this trial was adequately planned or what its observed power was. The registry does not report the original sample-size justification. Separating design information from retrospective illustrations prevents a planning calculation from being mistaken for evidence about the certainty of the completed trial.

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.

The registry record leaves important details about trial conduct and analysis unspecified. Although the study is described as randomized and single-masked, the identity of the masked parties, the allocation procedure, and allocation concealment are not reported. Recruitment procedures beyond the eligibility description and website invitation are not fully described in the record.

The primary endpoint has documented losses to follow-up in both arms, but the method used to handle those missing observations is not reported. The record does not state the primary analysis population, whether the analysis followed intention-to-treat principles, or whether sensitivity analyses assessed assumptions about missing outcomes. These details affect how confidently the one-month comparison can be interpreted.

The posted primary result includes means and arm-specific confidence intervals, but not a confidence interval for the between-group effect or a stated threshold for a meaningful change in decisional conflict. The record also does not report sufficient detail to reproduce the sponsor's analyses from participant-level data. The discrepancy between the six-month sponsor-reported test and the unadjusted recalculation cannot be resolved from the information recorded.

There are multiple secondary outcome areas and follow-up times, while multiplicity handling and prespecification are not reported. Numerical results are available for only the listed uptake comparisons in this analysis; results for the other named secondary outcomes are not reported in the registry record. As a result, conclusions about the broader effects of the decision tool on knowledge, regret, anxiety, or worry cannot be drawn from these posted results.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Incorporating Veterans' Preferences Into Lung Cancer Screening Decisions — NCT02899754.
• Registry record last updated 2024-11-14; status last verified 2024-11.
• Results first posted 2024-11-14.
• 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

NCT02899754 is a completed randomized, single-masked, parallel-group trial sponsored by the VA Office of Research and Development. It compared a web-based Lung Cancer Screening Decision Tool with general disease-prevention and health-promotion information unrelated to lung cancer in eligible veterans. The listed enrollment was 140.

The primary endpoint was decisional conflict one month after the intervention. The reported group means were 25.7 for the intervention arm and 29.9 for the control arm; the registry reports a two-sided t-test p-value of 0.18. The posted secondary uptake results show higher counts in the intervention arm at six and nine months, with registry-reported tests and separate unadjusted recalculations.

The record does not report the original sample-size justification, detailed missing-data methods, analysis populations, or multiplicity handling. Adverse-event results list zero affected participants in the serious and other categories in both arms, with zero deaths recorded.