A Retrospective Observational Study to Assess the Impact of Co-morbidities on Treatment Response in Inflammatory Bowel Disease: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Retrospective Observational Study to Assess the Impact of Co-morbidities on Treatment Response in Inflammatory Bowel Disease (NCT02861118), a registered phase not stated study. The primary outcome is Impact of the Comorbidities Profile in Inflammatory Bowel Disease (IBD) Participants on Lack of Treatment Response to Biological Therapy.

2. Clinical Question

The study's stated aim was to evaluate the impact of the comorbidities profile on treatment response in IBD participants receiving biological therapy. The primary analysis concerns an association between comorbidities and a response category, rather than a direct comparison of assigned treatments. The outcome is expressed as an odds ratio, with a 95% confidence interval specified as the measure of uncertainty. Neither the estimated odds ratio nor its confidence interval is reported in the registry record, so the direction, size, and precision of the primary association cannot be described from the record.

The registry lists secondary questions about extraintestinal manifestations in relation to lack or loss of response, the percentage of IBD participants with comorbidities, and comorbidity percentages within disease severity levels for Crohn's disease and ulcerative colitis. These broaden the descriptive and association questions, but do not establish that comorbidities caused a change in response.

3. Trial Design

VERNE is described as a retrospective observational study of comorbidities and treatment response in inflammatory bowel disease. The registry lists an enrollment of 310 and a completed status. The recorded start date is 2016-10-26 and the primary completion date is 2018-04-04. Takeda is listed as the sponsor.

The study has separate cohorts for Crohn's disease and ulcerative colitis. Cohort 1 includes participants with Crohn's disease who received biological treatment between June 2011 and June 2013. Cohort 2 includes participants with ulcerative colitis who received biological treatment during the same stated period. The registry's intervention entry is “No Intervention,” consistent with a study that observes treatment already received rather than assigning it as part of the study.

Allocation is not stated. The record does not describe randomization, and the cohorts are not a like-for-like control comparison. They represent different IBD diagnoses, with outcome definitions tailored to each disease. The registry's primary analysis is described as logistic regression adjusted for sociodemographic and clinical profile. Further details about covariate selection, model specification, or how the cohorts were analyzed are not reported in the registry record.

The primary outcome timeframe is up to 10 weeks after the start of biological treatment. The registry defines lack of response using change from baseline in HBI for Crohn's disease or PMS for ulcerative colitis after 10 weeks of anti-TNF treatment. The registry also lists secondary outcomes concerning extraintestinal manifestations, loss of response, and the distribution of comorbidities. It identifies 7 outcome measures and 9 analyses, but does not report the results for the primary or secondary analyses in the record.

The study's retrospective nature means that treatment and patient characteristics were observed in existing clinical circumstances. As a general statistical consideration, associations in such data can reflect differences among participants beyond the exposure being studied. Adjustment can account for measured characteristics included in a model, but does not by itself establish causality or eliminate potential effects of unmeasured factors.

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: Participants with a diagnosis of ulcerative colitis (UC) or Crohn's disease (CD) who started treatment with biologics between June 2011 and June 2013, participated in the study. Total 357 participants were registered, out of which only 310 participants were analyzed.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Analyzed participants included all participants who didn't had screen failure, met all the inclusion criteria and had enough information about response to the biological treatment.

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: Analyzed participants included all participants who didn't had screen failure, met all the inclusion criteria and had enough information about response to the biological treatment.

6. Primary Endpoint

The registry lists the primary outcome as Impact of the Comorbidities Profile in Inflammatory Bowel Disease (IBD) Participants on Lack of Treatment Response to Biological Therapy, measured over Up to 10 weeks after start of treatment with biologics.

The primary endpoint is an association between the comorbidities profile and lack of response to biological therapy in participants with inflammatory bowel disease. The stated effect measure is an odds ratio, and uncertainty is to be expressed with a 95% confidence interval. The timeframe is up to 10 weeks after treatment starts.

For Crohn's disease, the registry defines lack of response as a reduction of 2 points from baseline in the Harvey-Bradshaw Indices (HBI) score after 10 weeks of anti-TNF treatment. HBI includes general well-being, abdominal pain, number of liquid stools per day, abdominal mass, and complications. Its total is the sum of sub-scores; the registry describes scores <5 as remission, 5-7 as mild disease, 8-16 as moderate disease, and >16 as severe disease.

For ulcerative colitis, the definition uses the Partial Mayo score (PMS). PMS includes stool frequency, rectal bleeding, and physician's global assessment. The total is the sum of sub-scale scores, with 0 described as normal and 9 as severe disease. The registry's wording defines lack of response as a reduction of 2 points from baseline in PMS after 10 weeks.

These disease-specific measures provide a way to classify response using symptom and clinical assessment scores. General statistical context: converting a score change into a response category makes the outcome suitable for a logistic regression analysis, but the category depends on the chosen definition. The registry record does not report the primary odds ratio or its confidence interval, so the observed association and its precision cannot be assessed from the record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary analysis is described as logistic regression. General statistical explanation: logistic regression is commonly used when an outcome is represented by categories such as whether a specified event or response classification occurred. It estimates how predictors relate to the odds of the outcome, with estimates dependent on the model's coding, included predictors, and assumptions. In this study, the outcome of interest is lack of response, and the reported effect measure is an odds ratio. The registry states that models were adjusted for sociodemographic and clinical profile, but does not identify the exact covariates, coding decisions, or model-building procedure.

What does the odds ratio represent?

An odds ratio compares the odds of the modeled outcome across a specified contrast. It is not the same as a difference in response rates, and it should not be read as a probability or percentage change. To interpret its direction, the contrast and reference category must be known. The registry record does not provide the fitted estimate or all details needed to specify the modeled contrast. It is therefore not possible to translate the stated analysis into a numerical or directional finding.

General statistical explanation: adjustment means that a model accounts for selected variables while estimating the association of interest. This can help address confounding by measured characteristics when the model is appropriately specified and data support the estimation. It cannot guarantee that all sources of confounding have been removed. In a retrospective observational study, characteristics associated both with comorbidity and with treatment response may influence the observed association. The registry's broad statement that sociodemographic and clinical profile was adjusted for does not identify which factors were used or whether unmeasured confounding was considered.

What does the confidence interval add?

The 95% confidence interval is the stated uncertainty measure for the odds ratio. General statistical explanation: an interval expresses a range of estimates compatible with the data and the statistical procedure under its assumptions. Its width helps describe precision, and its position relative to the null value can inform interpretation when the estimate and interval are available. The registry record does not report the interval, so neither precision nor the interval's relation to the null can be evaluated.

The endpoint itself is based on disease-specific clinical scores. HBI is used for Crohn's disease and includes general well-being, abdominal pain, liquid stools per day, abdominal mass, and complications. The registry describes remission as HBI <5, mild disease as 5-7, moderate disease as 8-16, and severe disease as >16. PMS is used for ulcerative colitis and sums stool frequency, rectal bleeding, and physician's global assessment, with a total from 0 described as normal to 9 described as severe disease. The primary outcome definition is a reduction of 2 points from baseline after 10 weeks of anti-TNF treatment. Use of disease-specific scales reflects the distinct score systems named in the record; the registry does not report how score changes or incomplete score observations were handled in detail.

Why can over-dispersion matter?

General statistical explanation: over-dispersion occurs when observed variability is greater than a model expects. It can make standard errors too small if the model's variability assumptions are not appropriate, leading to confidence intervals that are too narrow. The registry does not report an over-dispersion assessment or identify it as an issue, and the stated primary method is logistic regression. No claim about over-dispersion or a particular correction can be made for this analysis.

The record lists 7 outcome measures and 9 analyses. General statistical explanation: when several outcomes or analyses are examined, the chance of obtaining at least one apparently notable result can increase if each result is interpreted independently. A multiplicity strategy may be used to address that issue, but such a strategy is not reported in the registry record. The number of outcomes and analyses alone does not establish that a correction was or was not applied.

A sample-size illustration is described as planning based on the observed effect, rather than as the study's own design calculation. The stated assumptions are a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect set to a stated fraction of the observed effect. The record also distinguishes illustrations for time-to-event outcomes, which use total events, from other outcomes, which use participants per arm. These are planning assumptions, not evidence of observed power. The registry record does not report a protocol sample-size justification, and the primary effect estimate needed to interpret such an illustration is not reported in the registry record.

General statistical explanation: a confidence interval and a planning calculation answer different questions. The interval describes uncertainty around an observed estimate under a specified analysis; a planning illustration describes the sample size associated with assumed conditions and a target effect. A planning result is not a measure of how likely the observed analysis was to detect an effect after the fact. No interim-analysis plan, detailed statistical analysis plan, or additional model diagnostics are reported in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The registry identifies 7 outcome measures and 9 analyses. Its primary outcome concerns the association between comorbidities and lack of response; listed secondary outcomes address extraintestinal manifestations in relation to lack or loss of response and comorbidity percentages overall and by disease severity. This number and range of analyses makes it relevant to distinguish a designated primary question from additional exploratory or descriptive questions. The registry record does not report whether a multiplicity adjustment, hierarchical testing strategy, or other approach to multiple comparisons was planned or used.

The record describes the analyzed population as participants who did not have screen failure, met all inclusion criteria, and had enough information about response to biological treatment. This gives a broad eligibility description for analysis, but does not specify how much information was required, how missing measurements were handled, or whether imputation or other missing-data methods were used. The number of participants excluded for missing or insufficient information is not reported in the registry record.

The registry enrollment is 310, while the safety entries separately list 194 participants at risk in the Crohn's disease cohort and 116 in the ulcerative colitis cohort. These figures describe different record fields; the registry does not explain whether analysis populations differed across outcomes, or provide outcome-specific denominators for the primary model. The primary analysis population is described by eligibility and data availability rather than by a reported outcome-specific count.

The registry does not report a detailed statistical analysis plan, interim analyses, or the method used to address missing data. It also does not specify whether all listed outcomes were analyzed in the same population or how any departures from planned analyses were handled. Without those details, the extent to which missingness or multiple testing could affect interpretation cannot be assessed from the registry record.

11. 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 5% reporting threshold.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: infusion related reaction (0.9%).

The registry's safety entries list serious adverse events and other adverse events separately for each disease cohort, with participants at risk shown for each category. In the Crohn's disease cohort, serious adverse events were recorded as 0 among 194 at risk, and other adverse events were recorded as 0 among 194 at risk. In the ulcerative colitis cohort, serious adverse events were recorded as 1 among 116 at risk, and other adverse events were recorded as 0 among 116 at risk. Deaths were recorded as 0 in each cohort.

The record also names “Infusion related reaction” as a top serious term and gives 0.9%. The registry record does not provide further context for that term, such as its event count or how it relates to the cohort-level serious-event entry. It should be read as a listed safety term and percentage, not as a complete account of the safety experience.

General statistical context: a zero recorded count does not prove that an event is impossible or that none occurred outside the study's observation and reporting procedures. Safety summaries depend on how events were defined, collected, and attributed, and the registry does not report detailed ascertainment methods or the precise adverse-event observation period. These figures describe the posted registry entries; they do not establish comparative safety because the cohorts are not a randomized, like-for-like control comparison.

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

The registry record establishes that VERNE examined comorbidity profiles in relation to treatment response among participants with inflammatory bowel disease who received biological therapy. It identifies a primary endpoint based on disease-specific clinical scores, an adjusted logistic regression approach, an odds ratio as the effect measure, and a 95% confidence interval as the uncertainty measure. These details describe the question and intended analysis, but they are not the numerical result.

The primary odds ratio and confidence interval are not reported in the registry record. Accordingly, the record does not establish whether comorbidity profile was associated with higher or lower odds of lack of response, how large any association was, or how precisely it was estimated. It also does not support a conclusion about statistical significance. The same restraint applies to secondary outcomes: their titles identify questions, but their numerical findings are not reported in the registry record.

Even if an adjusted association were reported, this retrospective observational design would not establish that comorbidities caused a change in treatment response. Participants were not assigned at random to a comorbidity profile, and allocation is not stated. Adjustment for sociodemographic and clinical profile can account for measured factors included in the analysis, but cannot rule out residual confounding or other limitations of observed clinical data.

The study's disease cohorts also use different response scales: HBI for Crohn's disease and PMS for ulcerative colitis. The registry defines lack of response through a reduction of 2 points from baseline after 10 weeks of treatment. Those endpoint definitions are specific to the record; they should not be confused with a direct comparison of Crohn's disease against ulcerative colitis or with a comparison of assigned therapies.

The safety entries report the event counts and at-risk figures recorded for each cohort, along with a serious-event term and percentage. They do not provide a controlled comparison of safety. Overall, the record is informative about the study's purpose, cohort structure, endpoint definition, and broad analysis method, but the missing outcome estimates limit what can be concluded about the observed associations.

13. Why This Trial Matters Statistically

VERNE addresses a statistical question that arises in clinical data: whether characteristics accompanying a disease are associated with response to treatment. Comorbidities may coexist with differences in sociodemographic and clinical profile, so the study's stated use of adjusted logistic regression is relevant to how such an association can be examined. The record also makes the target outcome explicit by defining lack of response with disease-specific scores.

For biostatisticians and clinical researchers, the distinction between a study question and a reported result is central. The registry identifies odds ratio and 95% confidence interval as the primary effect summary, but does not report either estimate. That gap illustrates why a complete interpretation requires effect estimates and uncertainty, not merely an endpoint title or a statement that a model was used. Without those results, readers cannot judge direction, magnitude, or precision.

The registry also helps highlight design considerations. This is a retrospective observational study with disease-based cohorts, not a randomized treatment comparison. A regression adjustment can be useful for measured covariates, but the extent of adjustment depends on which variables were collected, how they were modeled, and how missing information was handled. Those details are not reported in the registry record.

The study's listed secondary outcomes span response, loss of response, extraintestinal manifestations, and comorbidity prevalence across disease severity. This breadth matters when interpreting a set of analyses, since multiple questions can require clear distinctions between primary and secondary findings and transparent reporting of multiplicity procedures. The registry record identifies the number of outcome measures and analyses, but does not report a multiplicity plan or the numerical results.

Thus, the record is useful for understanding the intended analysis and the boundaries of what can be inferred. Its statistical relevance lies in the questions it asks and in the need to separate adjusted association, causal interpretation, and documented result.

14. 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 main limitation for interpreting the primary question is that its numerical result is not reported in the registry record. There is no odds ratio, confidence interval, or outcome count in the primary analysis description. The association's direction, magnitude, precision, and statistical significance therefore cannot be assessed.

The retrospective observational design limits causal interpretation. The registry states that the analysis was adjusted for sociodemographic and clinical profile, but does not name the covariates, describe their measurement, or explain their selection. The record also does not report how the comorbidity profile was defined or coded for the primary model. These omissions prevent assessment of residual confounding and model specification.

The cohorts are distinct disease groups, and the record does not identify a randomized or like-for-like control comparison. Crohn's disease and ulcerative colitis response are defined using different scores. The registry does not report whether the primary analysis was conducted separately by disease, jointly, or using a particular strategy to address differences between the score systems. Any direct comparison between the disease cohorts would therefore go beyond the information reported.

The analyzed population required adequate response information, but missing-data handling and outcome-specific analysis populations are not described. The registry does not report the number excluded for inadequate data, the frequency of missing score measurements, or whether imputation was used. It also does not state how multiplicity was addressed across the 7 outcome measures and 9 analyses, or provide a detailed statistical plan or interim-analysis information.

Finally, safety reporting is limited to the listed cohort counts, at-risk figures, deaths, and a top serious term. The registry does not describe event ascertainment, observation time, or detailed definitions. These record-specific limitations constrain assessment of both the primary association and the broader clinical context.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Retrospective Observational Study to Assess the Impact of Co-morbidities on Treatment Response in Inflammatory Bowel Disease — NCT02861118.
• Registry record last updated 2019-08-14; status last verified 2019-07.
• Results first posted 2019-08-14.
• 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.

18. Record Summary

NCT02861118, titled “A Retrospective Observational Study to Assess the Impact of Co-morbidities on Treatment Response in Inflammatory Bowel Disease,” is also identified as the VERNE study. Takeda is listed as sponsor, enrollment is 310, and the status is completed. The recorded start date is 2016-10-26; primary completion is listed as 2018-04-04.

The study describes Crohn's disease and ulcerative colitis cohorts who received biological treatment between June 2011 and June 2013. Its primary question concerns comorbidity profile and lack of response, analyzed using adjusted logistic regression with an odds ratio and 95% confidence interval. Response is defined with HBI for Crohn's disease and PMS for ulcerative colitis, up to 10 weeks after treatment starts.

The registry lists 7 outcome measures and 9 analyses. It does not report the primary odds ratio, confidence interval, or numerical results for the listed secondary outcomes. The posted information therefore describes the design and analysis question, but does not establish the observed association between comorbidity profile and treatment response.