HALO Patient Registry: Ablation of Barrett's Esophagus: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for HALO Patient Registry: Ablation of Barrett's Esophagus (NCT00848237), a na na study. The primary outcome is Endoscopic Clearance Rate for Barrett's Esophagus-Percentage of Patients With no Endoscopically Visible Barrett's Esophagus at 1 Year Follow-up.

2. Clinical Question

This clinical question is descriptive: it concerns the observed status of participants after treatment and follow-up, rather than a contrast between randomized treatments. Without a comparator group, the registry can describe outcomes among recorded participants but cannot by itself isolate the effect of radiofrequency ablation from other influences. The record’s objective was to provide participating physician investigators with a framework for collecting outcomes related to use of the HALO systems.

The endpoint is specifically based on endoscopic visibility. The endpoint description cautions that this is a visual measure and is not reliable or accurate; it identifies biopsy-based assessment as a better measure of clearance. That caveat matters when interpreting the clinical question: the measure addresses visible Barrett’s esophagus, not a biopsy-defined state. The registry record does not report a biopsy-based primary result, nor does it report a numerical one-year visual clearance result.

3. Trial Design

The HALO Patient Registry was a multicenter patient registry with prospective and retrospective components. It is listed as completed, with a start date of 2007-07 and primary completion date of 2014-07. The recorded enrollment is 5521. Its purpose was to support treatment and follow-up of people with Barrett’s esophagus and to collect outcomes related to use of the HALO Ablation Systems.

The registry is described as a single-group study. Radiofrequency ablation with the HALO systems is the intervention, and no separate control group is listed. Allocation is recorded as not applicable, and masking is listed as none. These design features mean that the primary outcome is a description of the treated registry population, not a randomized between-group estimate. The record does not describe a protocol-defined comparator or a blinded assessment process.

Eligibility language in the treatment description includes Barrett’s esophagus or intestinal metaplasia visible endoscopically or histologically. The broader registry description identifies non-dysplastic intestinal metaplasia, low-grade dysplasia and high-grade dysplasia. The record does not provide further eligibility criteria or explain how those categories were represented in the primary endpoint population.

The listed primary endpoint concerns absence of endoscopically visible Barrett’s esophagus at one-year follow-up. The population attached to this outcome is 4011 subjects who contributed answers. Although the registry lists six outcome measures, it lists no analyses. The primary endpoint percentage, its uncertainty and a breakdown of follow-up completeness are not reported in the registry record. A sample-size justification for the registry is not reported in the registry record; any planning illustrations should not be treated as evidence of the design’s observed precision or as observed power.

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

Analysis population for the primary outcome, as defined in the registry: 4011 subjects provided the answers

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 Endoscopic Clearance Rate for Barrett's Esophagus-Percentage of Patients With no Endoscopically Visible Barrett's Esophagus at 1 Year Follow-up, measured over 1 year.

The primary endpoint is the percentage of participants with no endoscopically visible Barrett’s esophagus at one-year follow-up. It captures a visual finding made during endoscopy: whether Barrett’s esophagus is apparent to the endoscopist at the specified follow-up. The record identifies 4011 subjects as having contributed answers for this outcome, but does not report the resulting percentage.

This type of endpoint is binary at the participant level: a participant either meets the stated visible-clearance definition or does not. Reporting a percentage can make the frequency of that status straightforward to understand. In general, a clearly defined time point also helps make the outcome interpretable, because the proportion is tied to a specified follow-up occasion rather than an unspecified period.

The registry’s own description adds an important qualification. It says the endpoint is visual and “not reliable or accurate,” and says biopsy-based assessment is a better measure of clearance. Thus, the endpoint should be understood as absence of visible Barrett’s esophagus, not as proof that tissue-level intestinal metaplasia is absent. The record does not report a numerical result for this visual measure or a biopsy-based result that could address the stated limitation.

As a general statistical point, a percentage alone does not convey how precisely it estimates an underlying frequency. Precision depends on the number of participants contributing outcome data and on the amount of uncertainty around the estimate. No confidence interval or other uncertainty estimate for this endpoint is reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

Registry design and statistical interpretation answer different questions. The design describes who was followed, what intervention was recorded and whether a comparison group was included. The statistical analysis would describe how outcomes were summarized and how uncertainty was quantified. The registry record identifies a single-group, multicenter registry and a binary primary outcome, but lists no analyses. No inferential method for the primary endpoint is reported in the registry record.

What does a binary percentage represent?

As a general explanation, a binary outcome assigns each participant to one of two categories according to a stated definition. For this endpoint, the definition is no endoscopically visible Barrett’s esophagus at one-year follow-up. The percentage would express the number meeting that definition relative to the number included in the outcome calculation. The registry identifies 4011 subjects who contributed answers, but the endpoint percentage itself is not reported. Without the numerator, the posted record does not allow the proportion to be reconstructed.

A proportion is descriptive. It does not by itself show whether an outcome is more common than it would have been under a different treatment or no treatment. For that causal question, a suitable comparison is needed, together with methods that account for the study design. This registry is single-group and has no listed allocation or control group, so its stated structure supports description of recorded outcomes rather than a randomized contrast.

What does uncertainty add?

As a general explanation, a confidence interval is an estimate of precision around a statistical estimate. A narrower interval indicates greater precision under the assumptions of the method; a wider interval indicates more uncertainty. An interval does not fix weaknesses in measurement, missing follow-up or study design. No confidence interval for the primary endpoint is reported in the registry record, so the precision of a primary percentage cannot be assessed from the posted result.

The number contributing outcome data is relevant to precision, but it does not determine precision by itself. The observed proportion also matters, as do the analysis method and assumptions. Since the registry record does not report the primary percentage, it is not possible to state a confidence interval or evaluate its width. No new estimate should be inferred from the enrollment and answer-population counts alone.

How should the design shape interpretation?

In general, random allocation helps make comparison groups similar on average with respect to measured and unmeasured characteristics. Here, allocation is listed as not applicable, masking as none, and the model as single group. There is no like-for-like control arm to serve as a counterfactual comparison. Thus, even if a numerical within-group outcome were available, it would remain a description rather than a randomized estimate of a treatment contrast.

Multicenter participation can draw on experience across more than one center, but it does not guarantee that participants or assessments are representative or consistent. The registry record does not report center-level results, the number of participating centers, or methods for accounting for variation between centers. These details cannot be assumed.

What do the planning assumptions mean?

As a general explanation, a sample-size calculation estimates how many observations a hypothetical study might need under specified assumptions. The registry’s planning description specifies a two-sided significance level of 0.05, equal allocation between two arms, a normal approximation, and an effect defined as a stated fraction of the observed effect. Those are assumptions for planning illustrations, not a description of the actual single-group registry’s allocation. The record also says that time-to-event illustrations use total events, while other illustrations use participants per arm.

These planning assumptions do not establish the registry’s actual sample-size justification, and that justification is not reported in the registry record. Planning outputs are not observed power. In particular, a planning calculation cannot substitute for an observed outcome estimate or its confidence interval. The record does not provide numerical planning outputs in the registry record described for this analysis, and none can be inferred.

Why does measurement matter statistically?

As a general point, statistical methods can summarize only the outcome that was measured. An endoscopic visual assessment and biopsy assessment are not identical measures. The registry explicitly cautions that the visual endpoint is not reliable or accurate and regards biopsy-based assessment as better. Even a precise estimate of the visual endpoint would not remove this measurement concern or convert the endpoint into a tissue-based measure.

Likewise, a well-defined endpoint cannot resolve missing-data concerns unless the analysis reports who contributed data and how unavailable observations were handled. The registry identifies 4011 subjects with answers, but does not report missing-data methods or the disposition of other enrolled participants. Therefore, the posted record does not allow assessment of whether the reported outcome population differs systematically from the full enrollment.

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("Treatment")
n      <- c(4011)
events <- c(2491)
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  Treatment  4011  2491  62.1  60.6  63.6  60.6  63.6  60.6  63.6

11. Multiplicity, Missing Data and Analysis Populations

The registry lists six outcome measures, but no analyses. It does not report whether one outcome was designated as primary for statistical testing, whether any other outcomes were exploratory, or whether a multiplicity adjustment was planned or applied. Multiplicity refers, in general, to the problem of interpreting results when several outcomes or hypotheses are examined: more opportunities to identify an apparently notable result can affect the overall error rate. No correction or testing strategy is reported in the registry record, so no claim about control of multiplicity is supported.

The record names the one-year visual clearance endpoint as primary and says 4011 subjects contributed answers. It does not give a data-flow account explaining how the 4011 were selected from the recorded enrollment of 5521. It also does not explain the amount, reasons or timing of unavailable outcome data, or whether missing observations were excluded, imputed or handled by another method. Those missing-data methods are not reported in the registry record.

As a general statistical matter, the way missing outcomes are handled can affect both the estimate and its uncertainty, especially if participants without recorded outcomes differ from those with recorded outcomes. The counts alone do not show whether such differences occurred. No analysis populations beyond the stated answer-contributing group, and no sensitivity analyses addressing missing data, are reported in the registry record.

The record also does not report interim analyses, stopping rules or a prespecified statistical analysis plan. Their existence or absence should not be inferred from the lack of listed analyses.

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.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: pain, nausea, vomitting, fullness, abdominal discomfort (0.1%), melena (0.1%), dysphagia and/or odynophagia (0.0%), esophagitis and/or ulceration (0.0%), fever (0.0%).

The safety record lists a treatment group with 5521 participants at risk for serious adverse events and 24 affected, and 5521 at risk for other adverse events and 42 affected. These are counts, not comparative estimates. The record does not identify a control group, and it does not report an analysis establishing whether events were caused by the intervention.

The listed serious-event terms and their reported percentages are: pain, nausea, vomitting, fullness, abdominal discomfort, 0.1%; melena, 0.1%; dysphagia and/or odynophagia, 0.0%; esophagitis and/or ulceration, 0.0%; fever, 0.0%; and stricture and/or dysphagia requiring dilatation, 0.0%. The spellings and values follow the registry listing. A displayed value of 0.0% should be read as the reported rounded value, not as proof that no event occurred.

Deaths are not reported in the registry record. The listed at-risk counts do not establish how completely adverse events were collected, and the record does not provide details of event definitions, follow-up duration for safety, or causality assessment. General statistical caution: event counts and percentages describe reports under the registry’s recording process; they do not by themselves show comparative risk or prove that an intervention caused an event.

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

The registry record documents a multicenter effort to collect treatment and follow-up outcomes for people with Barrett’s esophagus using the HALO Ablation Systems. It records enrollment of 5521 and describes a primary endpoint concerning absence of endoscopically visible Barrett’s esophagus at one-year follow-up. It also identifies 4011 subjects as having contributed answers for that endpoint. These details establish what the registry set out to observe, but the primary percentage is not reported in the registry record.

Because the numerical primary result is absent, the record does not establish how often participants met the visual endpoint. It also does not provide an interval or other statistic describing precision. No conclusion about the magnitude or consistency of the outcome can be drawn from the endpoint title or the answer-population count alone.

The single-group design does not supply a randomized comparison. In general, a within-group outcome cannot determine what would have happened to the same population under an alternative approach. The registry’s descriptive framework may record outcomes in routine practice, but the information in the record does not isolate a treatment effect or support a causal comparison.

The endpoint has an additional limitation acknowledged by the registry itself: it is a visual assessment, described as not reliable or accurate, with biopsy-based assessment identified as a better measure of clearance. Therefore, absence of visible Barrett’s esophagus should not be interpreted as a biopsy-confirmed result. The record does not report a numerical biopsy-based primary result.

The serious and other adverse-event counts are descriptive reports in the treatment group. Without a comparator, complete information about ascertainment and an analysis of attribution, they do not establish comparative safety or causation. Overall, the record is informative about registry structure, the stated outcome and recorded event counts, while leaving the primary numerical result and its uncertainty unreported.

14. Why This Trial Matters Statistically

This registry raises several useful statistical questions for clinical research. First, it distinguishes a registry’s purpose from a comparative trial’s purpose. The stated aim was to create a framework for participating physician investigators to collect outcomes related to use of the HALO Ablation Systems. A single-group registry can describe recorded experience, but its structure does not provide a randomized counterfactual comparison.

Second, it illustrates why endpoint definition and measurement quality matter. The primary endpoint is concrete in one sense: it specifies a visual finding at one-year follow-up. At the same time, the registry explicitly cautions that the measure is not reliable or accurate and says biopsy-based assessment is better. A precisely stated visual endpoint is not automatically a valid substitute for a different biological measurement.

Third, the record makes the distinction between enrollment and outcome contribution visible. It lists enrollment of 5521 and identifies 4011 subjects who contributed primary endpoint answers. Without a disposition of participants or missing-data methods, readers cannot determine how the outcome population relates to the full enrolled population. This is a reminder that a denominator and the process by which it was formed are central to interpreting a percentage.

Finally, statistical interpretation requires results as well as design information. The registry identifies its primary outcome, but the numerical result and its uncertainty are not reported, and no analyses are listed. A study can have a clearly described question without the public record supporting a numerical answer to that question. The safety counts similarly illustrate why descriptive event reporting should not be mistaken for a treatment comparison.

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 information recorded for this registry. The primary endpoint is based on endoscopic visibility, and its description explicitly calls it not reliable or accurate and identifies biopsy-based assessment as a better measure. The primary percentage and a biopsy-based outcome result are not reported in the registry record.

The outcome population is described as 4011 subjects who contributed answers, while recorded enrollment is 5521. The record does not explain the difference between these counts or report the completeness of one-year follow-up. Reasons for unavailable answers and methods for handling them are not reported in the registry record.

The registry is single group, with no listed allocation to a comparator and no masking. Consequently, the design does not offer a randomized like-for-like control comparison. The record lists six outcome measures but no analyses, and does not report confidence intervals, statistical tests or a detailed analysis plan. A sample-size justification is not reported in the registry record.

The adverse-event data provide event counts and listed terms, but the record does not report a comparison group or an analysis of causality. Deaths are not reported in the registry record. The safety information also does not describe event ascertainment or follow-up duration in the material recorded. These limitations constrain interpretation of both the primary endpoint and the safety counts; they do not establish that an unreported result is favorable or unfavorable.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: HALO Patient Registry: Ablation of Barrett's Esophagus — NCT00848237.
• Registry record last updated 2016-02-05; status last verified 2016-01.
• Results first posted 2016-02-05.
• 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

NCT00848237 is the completed HALO Patient Registry: Ablation of Barrett’s Esophagus, a prospective and retrospective multicenter registry. Recorded enrollment is 5521, with dates listed as 2007-07 for study start and 2014-07 for primary completion. The registry describes a single-group study of radiofrequency ablation using the HALO Ablation Systems, with no allocation and no masking.

The primary endpoint is the percentage of participants with no endoscopically visible Barrett’s esophagus at one-year follow-up. The endpoint population is described as 4011 subjects who contributed answers, but the percentage and its precision are not reported in the registry record. The record lists six outcome measures and no analyses. It records serious and other adverse-event counts, while deaths are not reported. The endpoint description cautions that visual assessment is not reliable or accurate and identifies biopsy-based assessment as a better measure.