Post-Market Surveillance With a Novel mHealth Platform: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Post-Market Surveillance With a Novel mHealth Platform (NCT03436082), a registered phase not stated study. The primary outcome is Number of Disease-Specific PROMs Completed.

2. Clinical Question

The study’s stated purpose was to pilot the feasibility of using a patient-led smartphone platform for real-world surveillance of outcomes. Its primary measure is a count of completed disease-specific PROMs, rather than a clinical effect estimate comparing the platform with another approach. PROMs were tailored to the procedure a participant received, so the bariatric and ablation groups followed different disease-specific questionnaires.

The registry record does not report the number of PROMs completed, the number completed at each scheduled time, or a summary of completion across participants. It therefore does not answer, from the posted results, how consistently participants completed the scheduled measures. Nor does it report a comparative estimate of outcomes between the two procedure groups. Those groups represent different patient and procedure contexts, not interchangeable arms in a treatment comparison.

3. Trial Design

This was a completed pilot study sponsored by Yale University. The registry lists a start date of 2018-01-26, a primary completion date of 2018-12-25, and enrollment of 60. The brief description frames the project as post-market surveillance using a patient-led smartphone-based mobile health platform after bariatric surgery or catheter-based atrial fibrillation ablation.

The registry describes two procedure-based groups: patients after sleeve gastrectomy or gastric bypass, and patients after catheter-based atrial fibrillation ablation. Each group was to pilot the use of Hugo for collecting information after the procedure. The intervention is classified as a device, described as a mobile health platform. The record does not describe a randomized allocation process; allocation is stated as not stated. It also cautions that the arms are not a like-for-like control comparison. The two groups differ by procedure and clinical context, so their results should not be treated as if they were assigned to competing platform strategies.

The primary outcome is the number of disease-specific PROMs completed over 8 weeks. The record says that questionnaires were emailed at enrollment, before the procedure, and at 1, 4, and 8 weeks post-procedure. It also describes recording the times when each questionnaire was emailed, initiated, and finally answered. These timestamps could characterize response timing, but the registry does not give numerical results for them.

The registry identifies 4 outcome measures and 2 analyses. It does not report the protocol’s sample-size justification, a formal statistical analysis plan, or a planned hypothesis test. The planning illustrations described for the results are not the study’s own design justification. Their assumptions include a two-sided significance level of 0.05, equal allocation between arms, a normal approximation, and effects expressed as fractions of the observed effect. These are planning illustrations, not evidence that the trial was designed or powered for a randomized comparison.

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: Assessed for # disease-specific PROMs completed

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: We enrolled a convenience sample of patients who were planning on undergoing either bariatric surgery (sleeve gastrectomy and gastric bypass) or catheter-based atrial fibrillation ablation at Yale-New Haven Hospital (YNHH) or the Mayo Clinic. 15 patients were enrolled for each procedure at each site, for a total of 60 patients.

6. Primary Endpoint

The registry lists the primary outcome as Number of Disease-Specific PROMs Completed, measured over 8 weeks.

The primary endpoint is the number of disease-specific patient-reported outcome measures completed. It is a count: each completed questionnaire contributes to the measure, with questionnaires tailored to the procedure received. The registry defines an 8-week timeframe and describes email invitations at enrollment, before the procedure, and at 1, 4, and 8 weeks after the procedure.

This endpoint is relevant to a feasibility pilot because the study aims to use a patient-led mobile platform for real-world outcome surveillance. Completion counts can indicate whether the planned questionnaire collection was carried out, while the recorded email, initiation, and final-response times can describe when responses occurred. These measures concern data collection and participant response, not whether a procedure or platform changes a clinical outcome.

The registry record does not report the completed-PROM count, completion at each scheduled point, or a participant-level completion summary. It also does not state how an incomplete questionnaire was defined, how partially answered questionnaires were handled, or whether all scheduled invitations were successfully sent. Without those details and the endpoint results, the primary measure cannot be quantified from the registry record. The described schedule and intended measurement are available, but the numerical findings needed to assess completion are not reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is a count, not a continuous measurement such as a score or a time-to-event endpoint. In general, count outcomes record how often an event or item occurs within a defined unit or period. Here the relevant items are completed disease-specific PROMs, and the stated observation period is 8 weeks. The interpretation of a count depends on knowing what qualifies as one completed measure, which participants are included, and whether every participant had the same opportunity to complete each scheduled measure. Those operational details are not reported in the registry record.

What does a count endpoint capture?

A count of completed questionnaires can describe the amount of response collected, but it is not automatically a proportion or a rate. A proportion would require both a numerator and a clearly defined denominator, such as the number completed relative to the number scheduled or expected. A rate would also require a defined exposure period or opportunity. The registry describes the planned email schedule but does not report the observed count or a denominator for completion. No rate or percentage should therefore be inferred from the description alone.

The scheduled administrations were at enrollment, before the procedure, and at 1, 4, and 8 weeks post-procedure. Since the PROMs were tailored to the procedure, the questionnaire content differed by procedure group. In general, repeated measurements from one person are related rather than independent: one participant’s response pattern can persist across follow-up occasions. If an analysis treats each completed questionnaire as an independent observation, it may understate that dependence. The registry does not report whether the analyses accounted for repeated observations, nor does it describe a model for the primary count.

Why can counts be more variable than a simple model expects?

In general, a basic count model may assume a particular relationship between a count’s average and its variance. Over-dispersion means the observed variability is greater than that assumption allows. It can arise when participants have different response tendencies, when opportunities differ, or when observations are clustered within participants. If over-dispersion is present and ignored, standard errors can be too small and inferential intervals or tests can appear more certain than they should. The registry record does not report a model, a dispersion assessment, or evidence about whether over-dispersion was evaluated. It is therefore not possible to say that a particular count model or correction was used.

Other approaches can be appropriate for count outcomes, depending on the question and data structure. General options include a model for the total count per participant, a model that accounts for repeated occasions, or a descriptive presentation of counts by scheduled time. Choosing among them requires decisions about the analysis unit, the expected count distribution, the handling of missing occasions, and whether group differences are being estimated. The registry record does not report these decisions. No particular approach should be attributed to the investigators based solely on the endpoint description.

What do the planning assumptions mean?

The planning illustrations are described as calculations based on the observed effect, rather than the trial’s own design. The listed assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. In general, a two-sided test allows departures in either direction under its alternative hypothesis, while equal allocation means assigning the same planned number of participants to each comparison group. A normal approximation uses a normal distribution to approximate the sampling behavior of an estimate or test statistic; its suitability depends on the data and design.

The illustrations consider effects stated as fractions of the observed effect. Such rows are planning scenarios about how many participants might be needed under the stated assumptions to detect an effect of a specified relative size. They do not show what was observed in the trial, and they are not the trial’s original sample-size rationale. The registry record does not report the protocol’s sample-size justification. These illustrations also do not establish observed power, which is not a useful replacement for interpreting an estimate and its uncertainty.

The registry describes two procedure-based groups but explicitly notes that they are not a like-for-like control comparison. Equal-allocation calculations do not change that design limitation. A planning formula can illustrate sample-size implications for an assumed comparison, but it cannot make distinct cohorts comparable or remove confounding. The record does not state that the arms were randomized, and allocation is not stated.

What can be inferred from the safety counts?

For safety, the record reports participants at risk and counts of serious adverse events, other adverse events, and deaths by group. These are descriptive counts. In general, a zero count means that no event of that category was recorded among the listed participants at risk during the study’s observation and reporting process. It does not establish that the underlying event probability is zero. Precision around a rare-event estimate depends on the number observed, the number followed, the duration and completeness of observation, and the reporting definitions. The registry record does not provide a confidence interval or further detail about event ascertainment.

In sum, the record gives the endpoint definition and some descriptive safety counts, but not the statistical model, uncertainty estimates, primary numerical result, or full analysis procedures. General statistical principles can clarify what those missing details would mean, but they cannot fill in unreported methods or results.

10. Multiplicity, Missing Data and Analysis Populations

The registry identifies 4 outcome measures and 2 analyses. The available record does not describe whether the analyses were prespecified, how outcomes were prioritized beyond identifying the primary endpoint, or whether any adjustment for multiple testing was planned. Multiplicity matters in general when several hypotheses are tested: as the number of tests grows, the chance of at least one apparently unusual result can increase if each test is interpreted separately. Adjustment methods and their purpose depend on the analysis plan. No such method is reported in the registry record, so no multiplicity procedure should be assumed.

The PROM schedule included enrollment before the procedure and follow-up at 1, 4, and 8 weeks. Repeated occasions create several possible ways to summarize completion, such as totals across the observation period or results by scheduled occasion. These are conceptually distinct summaries. The record does not state which summaries were analyzed, whether comparisons across time were planned, or how dependence among repeated observations was handled.

Missing-data methods are also not reported. The description says that disease-specific PROMs were emailed and that the email, initiation, and final-response times were available. It does not state how unanswered emails, started but unfinished forms, or missing items within a form were classified or analyzed. The number of participants eligible at each occasion and the number with observed data at each occasion are not reported for the primary endpoint. A completion count may be informative about participation, but without a stated denominator and missing-data rules it cannot fully describe response completeness.

The analysis populations are not defined in the registry record. It is not stated whether the primary analysis included every enrolled participant, only participants who initiated platform use, or another set. The relationship between the enrollment of 60 and the safety-group counts is not explained. These omissions limit reproducibility and make it difficult to determine exactly who contributed to each analysis.

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 0% reporting threshold.

The registry reports zero serious adverse events in the Bariatric group, with 29 participants at risk, and zero in the AF Ablation group, with 30 participants at risk. It also reports zero other adverse events and zero deaths in each group. No serious adverse event terms are listed.

These are observed counts in the registry record. They do not establish that the risk of an adverse event is zero. In general, an event that is not observed in a study may still occur outside the observed participants or observation period, and rare events can be difficult to characterize in a pilot. The record does not give confidence intervals, further detail on the observation period for safety, or information about how events were ascertained. The safety counts should therefore be read descriptively, without converting them into a claim of no risk.

The group counts are not a randomized comparison. They describe separate procedure populations, not participants assigned to competing strategies. The record does not explain how the safety-group counts relate to the enrollment figure of 60. These limitations prevent a causal safety comparison between the groups.

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

The study addresses whether a patient-led smartphone platform could be used for surveillance after bariatric surgery or catheter-based atrial fibrillation ablation. The registry describes a schedule for collecting disease-specific PROMs and indicates that timing information was available for emails and responses. This gives readers information about the intended data-collection process.

However, the primary outcome is the number of PROMs completed, and that result is not reported in the registry record. There is no numerical summary of completed measures, no account of completion by scheduled time, and no estimate of uncertainty. Consequently, the posted results do not quantify the level of PROM completion or the precision of such a quantity. The study’s stated feasibility purpose cannot be converted into a finding about how feasible the platform proved to be without the relevant outcome results.

The study also does not establish a clinical effect of the platform. The listed groups are based on different procedures and do not form a like-for-like control comparison. Allocation is not stated, and the record does not report an outcome comparison between the groups. The platform is part of the data-collection approach; the primary endpoint measures questionnaire completion rather than a clinical treatment benefit.

The safety counts are limited to the events reported in the record: zero serious adverse events, zero other adverse events, and zero deaths in both groups. Those counts do not demonstrate absence of risk, and they do not support a comparative safety claim. The record also lacks additional information about event ascertainment and uncertainty.

Thus, the defensible interpretation is bounded: the registry describes a completed pilot, its intended surveillance approach, and the PROM schedule, while leaving the primary numerical results and several analysis details unreported. This distinction separates a study’s objectives and measurement plan from evidence about its observed performance or effects.

13. Why This Trial Matters Statistically

This study is statistically notable because it puts the data-collection process itself at the center of a pilot. A platform intended for real-world surveillance depends on participants being able to receive, initiate, and complete repeated questionnaires. Recording when an email was sent and when a response was initiated or finalized can, in principle, help distinguish different points where collection may be delayed or incomplete. The registry describes those timestamps, though it does not report their results.

The primary endpoint is a count rather than a clinical outcome comparison. That choice aligns the stated measure with the pilot’s feasibility focus: it concerns completed PROMs. At the same time, a count is difficult to interpret without a clear denominator, specified rules for partial responses, and results by scheduled occasion. The record describes a multi-occasion schedule but does not report these details or the observed completion counts. This illustrates why endpoint definitions and reporting of the analysis population matter even when the measure is straightforward to state.

The procedure-based groups also highlight a core design issue. Bariatric surgery and atrial fibrillation ablation are distinct contexts with different tailored PROMs. Treating the groups as a randomized comparison would be misleading, and the registry explicitly says they are not like-for-like control arms. Statistical comparisons cannot overcome a design that does not create comparable groups without additional assumptions and information.

Finally, the registry’s planning illustrations should be distinguished from the study design. Assumptions about equal allocation, a normal approximation, and fractions of an observed effect can help illustrate hypothetical planning needs. They do not establish the rationale for the enrollment target, the precision of the study’s primary outcome, or the probability that a finding is true. With the primary numerical result and analysis details not reported, the record is most useful as a description of the pilot’s intent and data-collection framework rather than as a basis for an inferential conclusion.

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 most important limitation for interpreting the primary endpoint is that the registry record does not report its numerical result. The completed-PROM count, its distribution across participants, and completion at each scheduled occasion are not available. The endpoint definition also does not specify how partially completed questionnaires or missing items were handled.

The registry describes an enrollment of 60 but does not explain the relationship between enrollment and the safety counts of 29 in the Bariatric group and 30 in the AF Ablation group. It does not report analysis-population definitions, participant flow, or the number of participants contributing data at each follow-up occasion. This makes it difficult to establish which participants were represented in any particular analysis.

The two groups are based on different procedures and are explicitly not a like-for-like control comparison. Allocation is not stated, and no comparative analysis result is reported. Consequently, the record does not support causal comparisons between groups or a conclusion about platform effects relative to an alternative approach.

The registry lists 4 outcome measures and 2 analyses but does not describe the statistical analysis plan, multiplicity handling, model choice, or missing-data methods. It also does not report the protocol’s sample-size justification. The planning illustrations are not a substitute for those details. The record gives no confidence interval or p-value for the primary endpoint.

For safety, the listed counts are zero, but additional detail about event ascertainment, follow-up for safety, and uncertainty is not reported. As with any descriptive zero count, the result should not be read as proof that no risk exists. Taken together, these omissions limit assessment of precision, reproducibility, and generalizability from the posted record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Post-Market Surveillance With a Novel mHealth Platform — NCT03436082.
• Registry record last updated 2021-03-19; status last verified 2021-03.
• Results first posted 2021-03-19.
• 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

NCT03436082 is a completed Yale University pilot of a patient-led smartphone-based mobile health platform for surveillance after bariatric surgery or catheter-based atrial fibrillation ablation. The registry lists enrollment of 60, a start date of 2018-01-26, and a primary completion date of 2018-12-25.

The primary endpoint is the number of disease-specific PROMs completed over 8 weeks, with questionnaires scheduled at enrollment and 1, 4, and 8 weeks after the procedure. The record describes two procedure-based groups and cautions that they are not a like-for-like control comparison. It reports zero serious adverse events, zero other adverse events, and zero deaths in each listed group.

The registry does not report the numerical primary endpoint result, an analysis plan, or the methods used for missing data or multiplicity. It also does not state allocation or the protocol’s sample-size justification. The record therefore describes the pilot’s objective and measurement schedule but does not provide enough outcome information to quantify PROM completion or make a comparative effectiveness claim.