A Study to Assess the Compliance and Impact on Weight of Kurbo, a Pediatric Centered Weight Loss App: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Assess the Compliance and Impact on Weight of Kurbo, a Pediatric Centered Weight Loss App (NCT02880254), a randomized na study. The primary outcome is 3 Month Compliance With the Kurbo App as Well as the Kurbo App and PHC by a Group of Morbidly Obese Children and Adolescents in a Weight Management Program..

2. Clinical Question

Statistically, this is a pilot question about feasibility and the potential value of adding personal coaching to app use in a clinical weight management setting. It is not a direct test of the app against standard care alone, because both registered groups received standard of care. The record frames future estimation of compliance and BMI z-score differences between Kurbo and Kurbo plus PHC, using direct observation and modeling, but does not report those estimates.

3. Trial Design

The study is registered as randomized, single masked, and parallel. It enrolled 27 participants and is marked completed. The listed intervention groups are Kurbo only, described as app use plus standard of care, and Kurbo plus PHC, described as app use, personal health coach, and standard of care. Phase is listed as NA. The sponsor is Baystate Medical Center. The registry record lists the start as 2016-09 and primary completion as 2018-01-01.

A randomized design can, in principle, support a comparison between assigned groups by reducing systematic differences in measured and unmeasured characteristics. That interpretation depends on how randomization was generated and implemented, whether assignment was preserved, and how participants were analyzed. The random sequence method, allocation concealment, masking details, and analysis population are not reported in the registry record. The record says single masking but does not identify who was masked or how masking was maintained. These missing details limit assessment of potential selection, performance, and assessment biases.

The parallel structure means the registered groups are distinct treatment assignments during the study rather than a stated crossover sequence. The record does not report whether participants changed groups, how app or coach exposure was monitored, or whether adherence to assigned conditions was considered in the analysis. Nor does it report the duration of follow-up for the secondary BMI z-score measure beyond describing a 3 Month change. The primary compliance timeframe is 3 months.

This pilot study was intended to assess feasibility, compliance, and effects on BMI z-scores in children and adolescents in a weight management program. Its design therefore addresses use of the app as an adjunct to ongoing care, and the additional contribution of a personal health coach, rather than the effect of replacing clinical care. Although the record gives a total enrollment value and group-specific safety denominators, it does not report the outcome analysis population, participant flow, withdrawals, or the number contributing data to each endpoint. Those omissions make it difficult to determine how complete the outcome evidence would be even if results were available.

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: difficulty recruiting participants to be in the trial in the time frame; some agreed but then never came in to consent; there were no patients who consented but did not participate

Overall Study

Analysis population for the primary outcome, as defined in the registry: Assuming we find improved compliance and BMI-z scores, we will estimate compliance and BMI-z score differences between Kurbo and Kurbo + PHC. Estimates will be generated using direct observation as well as prediction through modeling.

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 had no patients who consented but did not participate in the study.

6. Primary Endpoint

The registry lists the primary outcome as 3 Month Compliance With the Kurbo App as Well as the Kurbo App and PHC by a Group of Morbidly Obese Children and Adolescents in a Weight Management Program., measured over 3 months.

The primary endpoint is described as 3 Month compliance with the Kurbo app, with or without a personal health coach, among children and adolescents in a weight management program. Its unit is percent compliant. The registry record says a questionnaire would be handed out to each subject every month, where applicable, and describes compliance as being assessed with a binomial probability test against a known compliance of 50%. The questionnaire text in the record is incomplete, and the operational definition of compliant status, including which responses or app-use behaviors count, is not reported in the registry record.

Compliance is a feasibility measure: it concerns whether participants engage with an intervention as intended, rather than directly measuring a health outcome. In an app-based intervention, engagement can help establish whether the intervention is usable in the setting and whether participants have enough exposure for later outcomes to be meaningfully evaluated. General statistical context: a clear compliance definition matters because a percentage has no consistent interpretation unless its numerator, denominator, observation window, and handling of missing responses are specified.

The registry also identifies 3 Month change in BMI z-score as a secondary outcome. BMI z-scores express BMI relative to a reference distribution, which can help describe weight status in growing children and adolescents. The record does not report the observed compliance percentages, the BMI z-score changes, or endpoint-specific denominators. Consequently, it does not show how the primary measure was distributed or how precisely any group comparison was estimated.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record includes a brief description of intended statistical procedures, but it does not report completed outcome analyses. The distinction matters: an analysis plan describes how investigators expected to evaluate data, while an analysis result requires observed data and a reported estimate or test. The record states that analyses were not reported, so the methods below explain the stated approach and the general statistical ideas it invokes without implying that results were obtained.

What does randomization contribute?

In general, random assignment aims to make the groups comparable at the start of a study, including with respect to characteristics that may affect outcomes. If allocation is implemented as intended and outcomes are analyzed appropriately, differences between assigned groups can be interpreted as evidence about the contrast between those assignments. Here, the registered contrast is app use plus standard of care versus app use with a personal health coach plus standard of care. Randomization does not create a standard-care-only comparison, since no such group is described. The specific randomization method and allocation procedures are not reported in the registry record.

The single-masked designation is also incomplete as a description of bias control. Masking can reduce the influence of expectations on behavior, care, or outcome assessment, depending on who is masked and what they know. The record does not identify the masked party or explain the masking procedures. Because the intervention includes an app and coaching, maintaining masking may be challenging, but whether it was feasible or attempted in particular ways is not reported in the registry record and should not be inferred.

How is compliance evaluated?

The primary endpoint is expressed as percent compliant over 3 months. The registry description says monthly questionnaires would be handed out where applicable and identifies a binomial probability test against known compliance of 50%. For a binomial test, observations are classified into two categories, such as meeting or not meeting a prespecified compliance definition, and the observed count is evaluated relative to a benchmark probability. General statistical context: this approach requires a clear definition of each trial or observation, the denominator, and assumptions about independence. Repeated monthly responses from one participant are not automatically independent observations, so the analysis must make clear whether compliance is defined per participant, per month, or in another way. The record does not report those details or the actual test result.

The protocol description specifies one-sided tests and a one-sided alpha of 0.025, stating that the interest is in whether app use results in better outcomes than usual care. Alpha is a prespecified threshold for interpreting evidence against a null hypothesis under a test’s assumptions; it is not the probability that a treatment works, nor does it describe the size or importance of an effect. A one-sided test evaluates evidence in a specified direction and does not use the same alternative hypothesis as a two-sided test. The record does not report the exact hypothesis for each comparison, whether the stated approach was followed, or any p-value.

What would an estimate and confidence interval add?

General statistical context: an estimate describes the observed magnitude and direction of a contrast, while a confidence interval conveys uncertainty around that estimate under the statistical model. A narrow interval indicates greater precision than a wide interval, but an interval is not a guarantee that the true effect falls within its bounds. For a compliance percentage, a confidence interval would help show how uncertain the estimated proportion is. For a comparison of groups, an interval for the difference would show the precision of that contrast. No such estimates or intervals are reported in the registry record, so effect size and precision cannot be assessed.

What does the BMI z-score outcome require?

The registry lists 3 Month change in BMI z-score as a secondary outcome. An analysis of change could describe within-group change, a difference in change between the groups, or an adjusted comparison of follow-up values. These approaches answer related but not identical questions. General statistical context: adjustment for baseline outcome values can sometimes improve precision and account for initial differences, while a direct change analysis depends on complete and comparable baseline and follow-up measurements. The registry record does not specify which model was used, what covariates were included, whether baseline values were adjusted for, or whether the analysis was completed.

For repeated measurements, modeling can account for correlation among observations from the same participant. The primary description mentions monthly questionnaires, but it does not state how repeated observations were handled. The population text says estimates of compliance and BMI z-score differences were intended to be generated through direct observation as well as prediction through modeling. The model form, predictors, assumptions, and role of the modeled estimates are not reported in the registry record. General statistical context: model-based estimates depend on the model specification and the assumptions used to connect observed data to unobserved or incomplete data.

How should the planning assumptions be read?

The registry’s sample-size assumptions describe planning illustrations based on the observed effect and a stated fraction of that effect. They specify a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation; non-time-to-event results are expressed as participants per arm. These are planning assumptions, not findings from the trial. Sample-size illustrations are not observed power, and they do not show that a study had a particular probability of detecting an effect after the data were collected. Their interpretation depends on the effect size and other assumptions used. The numerical illustrations and the observed effect are not reported in the registry record.

In general, over-dispersion occurs when observed variability is greater than the statistical model assumes. For binary or count outcomes, ignoring over-dispersion can make uncertainty appear too small and tests too optimistic. Whether over-dispersion was assessed or relevant to the planned compliance analysis is not reported in the registry record. Likewise, the record does not report diagnostic checks, model validation, or sensitivity analyses. Because no completed analysis is posted, the stated methods should be understood as a partial description of intent, not a basis for evaluating statistical execution.

10. Multiplicity, Missing Data and Analysis Populations

The registry identifies compliance as the primary outcome and BMI z-score change as a secondary outcome, with two outcome measures listed. The primary outcome description refers to tests for app use and app use with PHC, and the study’s broader aim includes assessing differences between these approaches. The registry record does not report whether multiple comparisons or multiple outcomes were handled with an adjustment, a prespecified hierarchy, or another strategy. Without that information, the chance of false-positive findings across a set of tests cannot be evaluated. No test results are posted, so there are no claims to assess for multiplicity in the reported analyses.

Missing-data procedures are not reported in the registry record. This is important for a questionnaire-based compliance endpoint because participants may have incomplete responses or may stop using the app. The record does not state how missing monthly questionnaires, absent app-use information, or missing follow-up BMI measurements would be handled. General statistical context: analyzing only participants with complete data can be biased if completeness is related to outcomes or group assignment; other approaches also require assumptions about why data are missing.

Analysis populations are not reported in the registry record. It is therefore unclear whether participants were intended to be analyzed according to randomized assignment, according to intervention actually received, or under another definition. Participant flow, withdrawals, and endpoint-specific denominators are also not reported. These details are needed to understand how enrollment relates to the data used for each estimate. The registry reports enrollment of 27, but that value alone does not establish the number included in any outcome 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 lists zero serious adverse events among 13 participants at risk in the Kurbo Only group and zero among 14 participants at risk in the Kurbo Plus PHC group. It also lists zero other adverse events in each group and zero deaths in each group. These are the recorded counts for the categories shown; they do not describe the nature, timing, duration, or severity of non-serious experiences beyond the category label.

General statistical context: an event count of zero does not establish that the underlying risk is zero. A safety estimate depends on how events were defined and collected, how long participants were observed, and whether all participants were assessed consistently. The registry record does not report those procedures, exposure time, a complete adverse-event narrative, or the method used to determine whether events were related to an intervention. Group-specific at-risk counts also do not establish how much follow-up each participant contributed.

Accordingly, these entries should be read as the safety counts recorded in the registry, not as a comprehensive statement about safety or as evidence that no adverse experiences occurred. They also do not provide a statistical comparison of safety between the groups.

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

The registry record establishes that a randomized, single-masked, parallel pilot study was completed, with app use plus standard of care compared with app use, personal health coach, and standard of care. It describes a primary focus on 3 Month app compliance and a secondary focus on change in BMI z-score. The record also gives group-specific adverse-event counts. These registry entries characterize the study’s purpose and registered design.

What the record does not establish is whether compliance was high or low, whether it differed between the registered groups, or whether BMI z-scores changed in either group. No outcome estimates, group contrasts, confidence intervals, or p-values are reported, and the registry lists zero analyses. Therefore, the posted information cannot quantify the direction, magnitude, or precision of an outcome effect. It also cannot show whether an apparent difference, if any, would be compatible with chance under a specified statistical model.

The registered comparison is specifically about adding a personal health coach to app use in the context of standard care. Both groups received standard of care, and the record does not describe a standard-care-only group. Thus, even if group results were available, this design would address the incremental contrast between app use alone and app use with PHC rather than isolating the app’s effect relative to standard care alone.

The study’s pilot purpose is relevant to interpretation. Pilot studies can help assess feasibility and inform later research, but the registry record does not report feasibility findings, recruitment performance, adherence estimates, or a sample-size justification. The enrollment value of 27 does not by itself indicate whether the study was adequately powered for any particular outcome. Planning illustrations based on observed effects are not observed power, and their numerical values are not reported in the registry record.

The safety counts are also limited. Zero recorded serious adverse events, other adverse events, and deaths in the listed groups are registry entries, not proof that no adverse experiences occurred. With event definitions, ascertainment, and follow-up details not reported, the counts cannot support a broad safety conclusion. Overall, the record is informative about the study question and registered framework, but it does not provide the results required to assess comparative effectiveness, precision, or a complete safety profile.

13. Why This Trial Matters Statistically

This study raises a statistical question that is distinct from simply asking whether an app is available: can participants engage with a pediatric-focused app, and does adding a coach change engagement or weight-related outcomes in a clinical program? Measuring compliance alongside BMI z-score change recognizes that an intervention’s practical use and its clinical outcomes are separate dimensions. A participant can engage without a detectable outcome change, and outcome patterns may be difficult to interpret if engagement is not measured clearly.

The randomized parallel comparison is relevant to the additional contribution of personal coaching because both registered groups received the app and standard of care, with coaching distinguishing the groups. This can frame an incremental comparison while avoiding an unsupported claim about app use versus standard care alone. The distinction matters when explaining what a study design can identify: conclusions must stay within the treatments actually compared.

The registry also highlights the importance of defining compliance. A percentage is interpretable only when the unit and denominator are clear, especially when questionnaires may be repeated monthly and app engagement can vary over time. Specifying how missing responses and repeated observations are handled is part of making a feasibility outcome reproducible and useful for planning future work. These definitions are not reported in the registry record.

From a statistical perspective, pilot studies can expose practical questions about measurement, engagement, and analysis that affect later evaluation. However, this record does not report outcome analyses, so it cannot demonstrate what the study learned about feasibility or the comparative outcomes. It supports a precise reading of the intended contrast and the limits of the available evidence, rather than a numerical conclusion about app compliance or BMI z-score change.

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 registry record has important limitations for evaluating this particular study. Although it identifies randomized assignment and single masking, it does not describe the randomization method, allocation concealment, or who was masked. It also does not report the actual amount of app use or coach contact, whether assigned interventions were followed, or how these exposures were recorded. These gaps restrict assessment of how the intervention contrast was implemented.

The primary compliance outcome is not fully operationalized in the record. It names a percent-compliant measure, monthly questionnaires, a 3-month timeframe, and a binomial test against 50%, but the questionnaire text is incomplete and the exact compliant threshold and denominator are not reported. It is also not clear how repeated monthly responses would contribute to a participant-level or group-level measure.

No outcome analyses or results are reported. The record does not provide estimates, uncertainty intervals, p-values, endpoint-specific denominators, participant flow, or analysis populations. Missing-data handling and the model for BMI z-score change are not reported. The status of completed therefore indicates study status in the registry, not that results are available in this record.

Generalizability is also difficult to assess because participant ages, baseline characteristics, recruitment details, and eligibility criteria are not reported. The study is described as involving children and adolescents with morbid obesity in a weight management program, but the record does not support a more detailed account of who participated. Finally, the safety entries lack event definitions, ascertainment methods, and exposure time. These record-specific omissions limit interpretation without establishing that particular procedures were absent from the conduct of the study.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: A Study to Assess the Compliance and Impact on Weight of Kurbo, a Pediatric Centered Weight Loss App — NCT02880254.
• Registry record last updated 2019-09-25; status last verified 2019-09.
• Results first posted 2019-09-25.
• 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

The registry record describes a completed randomized, single-masked, parallel pilot study of Kurbo in a pediatric weight management program. Enrollment is 27, and the registered groups received either Kurbo plus standard of care or Kurbo, a personal health coach, and standard of care. The primary outcome is 3 Month app compliance; BMI z-score change is listed as a secondary outcome.

The record describes a planned binomial test of compliance against 50%, with one-sided alpha of 0.025, but reports no outcome analyses or results. No compliance estimates, BMI z-score findings, statistical contrasts, confidence intervals, or p-values are reported. The safety entries list zero serious adverse events, zero other adverse events, and zero deaths in each group. These counts do not establish a complete safety profile. The registry therefore describes the study question and design but does not provide findings that resolve the comparative outcome questions.