Communicating Risk Information for Obesity: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Communicating Risk Information for Obesity (NCT01355224), a randomized na study. The primary outcome is Intentions to Lose Weight - Effect of Study Arm on Outcomes.

2. Clinical Question

3. Trial Design

The registry describes a randomized, double-masked, factorial study, with enrollment of 798 and status listed as completed. The recorded start was 2011-09 and primary completion was 2013-09. Boston University is listed as the sponsor. The phase is recorded as NA.

The factorial arrangement involved genetic and lifestyle feedback. The four conditions were no risk feedback, genetic risk feedback alone, lifestyle risk feedback alone, and genetic and lifestyle risk feedback together. This design distinguishes feedback content across arms, but the reported primary analysis is a regression of outcomes by study arm. The registry does not report the factorial model specification or separate estimates of genetic and lifestyle factor effects.

The study population is described as individuals recruited to the Coriell Personalized Medicine Collaborative (CPMC). The record’s summary says the study would examine attitudes and beliefs about obesity, health behaviors, and weight outcomes, and the feasibility of using genetic information to motivate efforts to reduce obesity. More detailed eligibility criteria, participant characteristics, and recruitment procedures are not reported in the registry record.

The primary outcome was intention to lose 10lbs in the next 6 months, measured on a 5-point Likert scale. It was assessed three months after viewing the obesity risk assessment. Arm-level means and 95% confidence intervals are reported. A linear regression analysis is also reported for the four study arms, with p = .0150. The regression coefficient, confidence interval for an arm contrast, covariate adjustment, and detailed model specification 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

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: Based on participants who completed the baseline survey

6. Primary Endpoint

The registry lists the primary outcome as Intentions to Lose Weight - Effect of Study Arm on Outcomes, measured over 3 months after viewing obesity risk assessment.

The primary endpoint measures a stated intention, not weight loss itself. Participants rated their intention to lose 10lbs in the next 6 months on a 5-point Likert scale, where 1 indicated strongly disagree and 5 indicated strongly agree. The assessment took place three months after participants viewed an obesity risk assessment.

An intention measure can capture a response to information before a behavior or weight outcome is observed. In this study, it addresses whether responses differed by the kind of risk feedback participants received. That makes it relevant to the stated research aim of examining attitudes, beliefs, and potential motivation. It remains a self-reported scale outcome: it does not establish whether a participant attempted weight loss, changed health behaviors, or lost weight.

The registry reports means and 95% confidence intervals by arm. The mean describes the average scale response within an arm, while the confidence interval expresses uncertainty around the estimated mean under the analysis assumptions. The measure’s interpretation depends on the scale and the timing of assessment; it should not be treated as a direct measure of obesity or a clinical outcome. The registry does not report a detailed explanation of how responses were handled if a participant did not complete the scale.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does an arm mean represent?

A mean is the average response across the observations included in an arm’s analysis. For this continuous scale outcome, the arm means describe the average stated intention to lose 10lbs in the next 6 months. They preserve the scale’s original units, rather than expressing an estimated change in weight. The registry lists the means alongside 95% confidence intervals, which communicate uncertainty in each estimated mean under the method’s assumptions. A confidence interval is not a range containing a known proportion of individual responses; it is an interval estimate for the group quantity.

The reported arm summaries are descriptive. A difference between two displayed means is not, by itself, a complete statistical comparison. A formal contrast would specify which arms are being compared and provide an estimate and uncertainty for that contrast. Those pairwise estimates are not reported in the registry record. The posted analysis instead describes a linear regression across the four arms.

What does the regression result tell the reader?

Linear regression models the relationship between an outcome and one or more explanatory variables. When study arm is the explanatory variable, the model can assess whether the outcome varies across arms, depending on how arm is represented and what other terms are included. The registry identifies the method as linear regression and reports p = .0150 for the analysis covering the four groups. A p-value measures how compatible the observed data are with a specified null model, under the assumptions of that analysis. It is not the probability that the null hypothesis is true, and it does not quantify the size or practical importance of an effect.

The record does not report the regression coefficients, standard errors, confidence intervals for contrasts, covariates, reference category, or coding of the factorial factors. Without those details, the reported p-value cannot be translated into an estimated contrast for a particular feedback condition against no feedback. The result also does not establish which pair of arms differs. No additional regression details should be inferred from the method label alone.

Why do confidence intervals matter?

A confidence interval gives a range of values compatible with the data and the method’s assumptions for the quantity being estimated. It helps convey precision: a narrower interval indicates less uncertainty about that estimate than a wider interval, all else equal. The arm intervals concern means, whereas a contrast interval would concern a difference or other comparison between arms. They answer different questions. Overlap or non-overlap between separate arm intervals is not a substitute for a directly estimated contrast and its uncertainty.

The record’s interval for genetic risk feedback is 2.24 to 3.5. It should be read as recorded. The registry does not explain its width or provide a separate corrected value. The other reported intervals are 2.93 to 3.19 for no risk feedback, 3.13 to 3.39 for lifestyle risk feedback, and 3.13 to 3.39 for genetic and lifestyle risk feedback.

How does the factorial arrangement relate to the analysis?

A factorial design organizes conditions around more than one intervention factor. Here, the intervention descriptions concern genetic and lifestyle feedback, and the arms include each type alone, both together, and neither. In general, factorial analysis can estimate effects associated with each factor and may assess whether their combination changes the relationship. The registry, however, identifies the reported method only as linear regression across the arms. Factor-specific estimates, an interaction estimate, and the model’s precise specification are not reported in the registry record. The design label alone does not establish which effects were estimated.

What is not conveyed by the sample-size illustrations?

Any sample-size figures framed as planning illustrations are hypothetical design calculations based on an observed effect and stated assumptions. Such calculations are not observed power for the completed study, and they do not document the trial’s original design justification. The registry record does not report the original sample-size justification. A planning illustration should not be interpreted as a new result or as evidence that the study had a particular probability of detecting an effect.

General statistical interpretation also depends on assumptions about the model and the observations. The record does not provide enough detail to assess the regression diagnostics, handling of missing responses, or whether the analysis accounted for features of the factorial design beyond treating arm as a predictor. These omissions limit how far the reported test can be interpreted, even though the method name and overall p-value are stated.

10. Analyses Reported by the Sponsor

The registry’s reported analysis is a linear regression covering the four study arms, with p = .0150. It does not include a regression estimate, confidence interval for an arm contrast, or enough model detail to reconstruct the sponsor’s analysis. The arm means and confidence intervals are descriptive summaries; they do not supply the missing regression contrast.

No recalculated arm-to-arm estimate is reported in the registry record. The record notes that values were missing or not estimable for a two-arm comparison of this continuous endpoint. Therefore, there is no recalculated contrast to set against the sponsor’s regression result. The sponsor’s reported overall analysis can be described, but its effect size and specific contributing arm comparisons cannot be recovered from the information recorded. The record does not identify a discrepancy between a sponsor analysis and a recalculation.

11. Multiplicity, Missing Data and Analysis Populations

The registry records three outcome measures and three analyses, but it does not provide enough detail to determine how the primary outcome relates to every listed measure or analysis. It records no posted secondary outcomes. The analysis reported for the primary outcome is a linear regression across the four study arms. The registry does not state whether other outcome analyses were planned, whether the reported analysis was prespecified, or whether any adjustment for multiple testing was used.

Multiplicity matters when researchers conduct several statistical tests: the chance of at least one apparently unusual result can rise as the number of tests increases. Adjustments and interpretation strategies depend on the analysis plan and the family of tests being considered. Because that plan and any multiplicity procedure are not reported in the registry record, the p-value should be read as the reported result of its named analysis, not as evidence that multiplicity was or was not addressed.

The record also does not describe the analysis population for the primary endpoint. It gives arm-level sample sizes for the outcome summaries, but does not explain whether all randomized participants were included, how missing responses were handled, or whether any exclusions occurred. The record’s enrollment figure should not be assumed to be the denominator for each analysis. Reasons for missing data, imputation methods, and sensitivity analyses are not reported in the registry record.

These gaps affect reproducibility and interpretation. Analysis populations and missing-data procedures can influence which observations contribute to estimates. The absence of reported detail does not establish that a particular method was or was not used; it means that the method cannot be determined from the registry record. The reported summaries and test are therefore best interpreted without assumptions about unreported analysis choices.

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 0% reporting threshold.

The registry lists serious adverse events and other adverse events by arm, with zero affected participants recorded in each listed category. The at-risk counts are 198 for no risk feedback, 203 for genetic risk feedback, 196 for lifestyle risk feedback, and 201 for both genetic and lifestyle risk feedback. The record does not list serious adverse event terms.

These entries describe the safety information recorded in the registry. They should not be read as proof that no event of any kind occurred: the figures are limited to the categories, definitions, reporting, and observation captured by the record. Death counts are not reported in the registry record. The safety counts are separate from the primary intention outcome and do not add information about the effect of risk feedback on the Likert-scale response.

The registry does not explain the relationship between the safety at-risk counts and the analysis populations for the primary outcome. It also does not provide further detail about event collection or follow-up in the safety entry. The recorded zeros can be stated as reported, but broader conclusions about safety or unrecorded events are not supported by these figures alone.

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

The main statistical finding recorded for the primary outcome is an overall linear regression result across the four study arms, with p = .0150. The arm means describe average stated intention at the assessment conducted three months after viewing the obesity risk assessment. Together, these results concern a self-reported intention measure in this study population and at this time point.

The result does not establish how large an arm effect was, because the regression estimate and its confidence interval are not reported. It does not identify which specific feedback condition differs from no risk feedback, or whether genetic feedback, lifestyle feedback, or their combination accounts for the overall result. Those conclusions require estimates and uncertainty for the relevant contrasts, which are not reported in the registry record.

Nor does an intention rating establish a change in actual behavior or weight. The registry’s broader aims include attitudes, beliefs, health behaviors, and weight outcomes, but the primary result is intention to lose 10lbs in the next 6 months. It should not be generalized into a claim about later weight outcomes or lasting motivation. Further outcome details and analysis results are not reported in the registry record.

A statistically notable p-value is not a measure of the magnitude, importance, or durability of an outcome. Its interpretation depends on the model and the comparisons represented in that model. The record does not describe covariate adjustment, the factorial model specification, missing-data handling, or multiplicity procedures. These omissions counsel against reading more into the result than the reported overall association between study arm and the intention measure.

14. Why This Trial Matters Statistically

Statistically, this study raises a question about how information is communicated and how outcomes are measured. The randomized factorial structure separates genetic and lifestyle risk feedback into distinct conditions, including a combined-feedback condition and a no-feedback control. That structure can support comparisons of feedback strategies, while the reported results require careful attention to which contrasts were actually estimated.

The primary outcome is a stated intention rather than a direct measure of weight or behavior. Keeping that distinction clear is important when interpreting results about communication and motivation. An intention scale can describe how participants respond to information, but it does not by itself establish what they later do. The follow-up timing is specified as three months after viewing the obesity risk assessment, which anchors the result to a particular assessment point rather than a long-term outcome.

The study also illustrates the difference between an overall test and estimates of specific comparisons. The registry reports a regression p-value across arms but not the corresponding coefficients or pairwise contrasts. Without those estimates, the overall test cannot answer every question readers might ask about individual feedback conditions. Clear reporting of contrast estimates, uncertainty, model specification, analysis populations, and missing-data procedures would make the statistical interpretation more complete. Those details are not reported in the registry record.

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 primary endpoint captures intention on a 5-point Likert scale, not observed weight loss or a verified behavior change. Its wording concerns intention to lose 10lbs in the next 6 months, measured three months after viewing the risk assessment. The registry does not report whether the scale was validated, how it was scored beyond the stated response anchors, or how incomplete responses were handled.

The genetic risk feedback arm has a recorded 95% confidence interval of 2.24 to 3.5. The registry does not explain this interval or provide an alternate value. Interpretation should retain the posted value rather than silently treating it as a transcription error. The record also lacks the regression estimate and contrast intervals needed to explain the overall test in terms of arm-specific effects.

Important design and analysis details are not reported in the registry record, including eligibility criteria, participant characteristics, the detailed factorial model, covariate adjustment, analysis populations, missing-data procedures, and multiplicity handling. The record does not give the original sample-size justification. These gaps limit assessment of precision, generalizability, and the robustness of the statistical analysis.

The registry describes the broader aims as including attitudes, beliefs, health behaviors, and weight outcomes, but the posted primary result is an intention measure. Results for further outcomes and details about longer-term follow-up are not reported in the registry record. The safety listing also does not provide event terms or death counts. These limitations mean that the record supports a focused description of the reported intention outcome, not a broader conclusion about behavior, weight, or safety.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Communicating Risk Information for Obesity — NCT01355224.
• Registry record last updated 2017-03-14; status last verified 2017-01.
• Results first posted 2017-03-14.
• 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

NCT01355224 is a completed randomized, double-masked, factorial study sponsored by Boston University. Enrollment was 798. The study compared no obesity risk feedback with genetic feedback, lifestyle feedback, or both kinds of feedback among individuals recruited to the Coriell Personalized Medicine Collaborative (CPMC).

The primary outcome was intention to lose 10lbs in the next 6 months, measured on a 5-point Likert scale three months after viewing an obesity risk assessment. The registry reports arm means and 95% confidence intervals and an overall linear regression result of p = .0150. It does not report a regression estimate or pairwise arm contrasts, and details of the model, analysis populations, missing-data methods, and multiplicity handling are not reported in the registry record.