The University of Michigan PCOS Intervention Using Nutritional Ketosis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The University of Michigan PCOS Intervention Using Nutritional Ketosis (NCT03987854), a na phase 1, phase 2 study. The primary outcome is Mean Percent Body Weight Loss.

2. Clinical Question

This question is descriptive of change within the study group. It is not, on the design information recorded, a direct estimate of how the program compares with another diet, usual care, or no intervention. Without a control group, changes over time cannot by themselves separate a program-related change from other influences, including changes that might occur over time independently of the program.

The registry record identifies HbA1c change as a secondary outcome. It lists 8 outcome measures and 8 analyses, but the additional outcome definitions and results are not reported in the registry record. The primary endpoint is mean percent body weight loss at 4 months. No primary outcome estimate, standard deviation, confidence interval, or test result is reported in the registry record, so the direction, magnitude, and precision of the observed change cannot be described.

3. Trial Design

The University of Michigan PCOS Intervention Using Nutritional Ketosis is recorded as a Phase 1 and Phase 2 study and as completed. The registry dates are 2019-07-05 for study start and 2020-03-01 for primary completion. Enrollment is listed as 29. These are registry design and enrollment details; they do not establish how many participants contributed data to the primary endpoint.

The design is listed as single-group, with allocation not applicable and no masking. The registry record describes a behavioral intervention called a diet and lifestyle program. Its brief summary says the goal was to pilot test an existing very-low carbohydrate diet intervention adapted for women with PCOS. The record does not describe a separate control arm, and it does not present the study as a randomized comparison. The listed arm is Complete Diet and Lifestyle Program.

A single-group design can describe outcomes observed among people who took part, but it does not directly estimate a contrast with an alternative. In a randomized comparison, assignment to different groups can support an estimate of the difference between those groups, subject to the design and analysis. No such group contrast is available in the design details recorded for this study. A before-and-after change within one group answers a narrower question: how the measured outcome changed over time among participants with available measurements.

The registry record does not report eligibility criteria, recruitment procedures, setting, program delivery details, adherence, co-interventions, or the number of participants with measurements at each time point. It also does not report a protocol, a statistical analysis plan, or a sample-size justification for the trial itself. Those omissions limit assessment of who the results would describe, how the intervention was implemented, and how planned analyses compare with analyses actually conducted.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: all participants at baseline

6. Primary Endpoint

The registry lists the primary outcome as Mean Percent Body Weight Loss, measured over 4 months.

The primary endpoint is mean percent body weight loss at 4 months. The registry defines it as the change from baseline weight to weight at 4 months, divided by baseline weight and multiplied by 100. The listed unit is percentage of weight. The endpoint is a continuous measure: each participant can have a measured value, and the reported group summary is intended to describe the mean of those values.

A percentage change expresses weight change in relation to the participant’s baseline weight, rather than reporting an absolute change in weight units. This can make changes interpretable relative to starting weight. The registry’s formula uses weight at 4 months minus baseline weight, so a decrease in weight would yield a negative value under that formula, even though the endpoint title uses the phrase weight loss. The registry record does not explain how this sign convention is handled in any reported result because the result itself is not reported.

For a continuous endpoint, the mean describes the group’s average, while a standard deviation describes how spread out participant values are around that mean. The endpoint record identifies the parameter as a mean and dispersion as a standard deviation. Neither the observed mean nor the standard deviation is reported in the registry record. The number of participants contributing endpoint measurements is also not reported there.

Measuring the endpoint at a specified follow-up time provides a defined point for describing change. However, the record does not report details about the timing window, measurement procedures, or handling of missing follow-up weights. Without the numerical result and those analysis details, the endpoint’s observed magnitude and precision cannot be evaluated.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the mean represent?

For a continuous outcome, the mean is the arithmetic average of the measured values in the analysis group. In this study, the designated primary measure is mean percent body weight loss at 4 months. The mean would summarize the average endpoint value among participants included in that analysis, but it would not show whether individual participants had similar or very different changes. The registry record identifies the standard deviation as the measure of dispersion intended to accompany the mean. Neither numerical summary is reported in the registry record.

General: A standard deviation describes the spread of observations around their mean. A smaller spread indicates that observations are more tightly clustered around the average; a larger spread indicates more variability. It is not the same as the uncertainty in the estimated mean. Uncertainty in the mean is commonly described using a confidence interval, but no confidence interval is reported for this endpoint in the registry record.

How should a confidence interval be understood?

General: A confidence interval gives a range of values compatible with the data and the assumptions of the method used to calculate it. Its width reflects precision: a narrower interval indicates greater precision, while a wider interval indicates greater uncertainty. The interval does not remove limitations in the study design, and it cannot turn a single-group estimate into a comparison with an unobserved control group. No interval for the primary endpoint is reported in the registry record, so its precision cannot be assessed.

What would a test result add?

General: A statistical test evaluates how compatible the observed data are with a specified null hypothesis under a model. A p-value is not the probability that a treatment works, nor does it measure the size or practical importance of an effect. For a within-group outcome, a test might address whether an average change differs from a chosen reference value, depending on the planned analysis. The registry record does not report a primary-outcome p-value, the null hypothesis, or the test used. No test result should be inferred from the endpoint description.

What does the record say about the planning illustration?

The stated planning illustration uses a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It describes planning to detect a stated fraction of the observed effect. The record explicitly distinguishes this illustration from the trial’s own design: this trial is listed as single-group, whereas the illustration assumes two arms. It is therefore not a description of the study’s observed analysis or evidence that the trial had a particular power.

The record also distinguishes the type of quantity used in planning: the planning output is framed as participants per arm for ordinary outcomes, while time-to-event planning is framed as a total number of events. The primary endpoint in this study is continuous, not time-to-event. No numerical planning outputs are included in the registry record details, so no sample-size value can be stated. The observed effect needed as an input to the illustration is also not reported for the primary endpoint. Such planning illustrations, even when available, are design scenarios rather than observed findings.

Why does over-dispersion matter?

General: Over-dispersion means that data vary more than a statistical model expects. When a model underestimates variability, standard errors may be too small and confidence intervals too narrow, making results appear more precise than they are. Appropriate modeling depends on the outcome type and data structure. The primary endpoint is described as a continuous measure summarized by a mean and standard deviation. The registry record does not report model diagnostics or adjustments for over-dispersion, and the record does not provide enough information to determine whether such an issue arose.

What analysis details are missing?

The registry record does not report the statistical test or model for the primary endpoint, whether the analysis used individual change scores or another approach to baseline and follow-up values, or whether estimates were adjusted for any covariates. It does not report an analysis population definition or how missing observations were treated. These details matter because different choices can affect the estimate, its uncertainty, and the participants to whom it applies.

General statistical interpretation also depends on the study design. In a single-group study, an estimated change over time does not isolate the effect of the intervention from changes associated with time, participant selection, measurement, or other influences. Statistical precision describes uncertainty under the analysis assumptions; it does not resolve the absence of a comparison group. Since the primary estimate and its uncertainty are not reported, neither the numerical result nor the adequacy of its precision can be evaluated from the registry record.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Complete Diet and Lifestyle Program")
n   <- c(21)
m   <- c(-7.67)
sp  <- c(6.1)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Complete Diet and Lifestyle Program  21  -7.670  6.100  -10.447  -4.893

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists 8 outcome measures and 8 analyses. It identifies mean percent body weight loss at 4 months as the primary endpoint and HbA1c change as a secondary outcome. The record does not provide the full set of outcome definitions and results, the analysis hierarchy, or the statistical plan. It does not report whether analyses were prespecified, whether one or more outcomes were treated as confirmatory, or whether a procedure was used to account for multiple testing.

General: When many outcomes or hypotheses are examined, the chance of at least one apparently notable result can increase if each is interpreted without regard to the others. Multiplicity procedures can address this issue, but the appropriate approach depends on the aims and planned hierarchy of the analyses. The count of listed outcomes and analyses does not, on its own, reveal the number of independent statistical tests or establish whether adjustment was needed. The multiplicity approach is not reported in the registry record.

The registry record also does not describe missing-data handling for the primary endpoint or other outcomes. It does not state how many participants had measurements at baseline and 4 months, whether participants with incomplete measurements were excluded, or whether any imputation or other method was used. The enrollment figure should not be treated as the endpoint analysis size.

Analysis populations are not defined in the record. For example, it does not state whether the primary analysis included all enrolled participants, only those with available measurements, or a different group defined by the analysis plan. These omissions prevent assessment of how missingness and analysis-population choices may have influenced the reported—or unreported—outcome estimates.

12. 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.

Most frequent non-serious adverse event terms

The registry record lists 0 serious adverse events among 29 participants at risk in the Complete Diet and Lifestyle Program group. It lists 3 other adverse events among 29 at risk and 0 deaths. These are counts as recorded, not rates calculated from the counts. The record does not list serious adverse event terms.

A serious-event count and an other-event count refer to different categories in the registry record. The entries do not describe the nature, timing, severity, or attribution of the 3 other adverse events. Those details are not reported in the registry record. The listed at-risk figures are not necessarily the number of participants assessed for every outcome over the full study period, and the record does not explain the event-monitoring procedures.

General: A zero count means that no event was recorded in the relevant category for the group and reporting period represented in the registry. It does not prove that the event cannot occur, and it does not establish that a program is free of risk. A comparison of safety between treatments would require a suitable comparator and sufficiently detailed, consistently collected safety data. No comparator group is listed for this study, so these counts should be read as descriptive rather than comparative.

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

The registry record describes a completed Phase 1 and Phase 2, single-group pilot of a diet and lifestyle program for women with PCOS. It defines a primary endpoint—mean percent body weight loss at 4 months—and records safety counts for the program group. This information identifies the study question and some aspects of its design, but it does not include the numerical primary endpoint result.

Because the mean and standard deviation are not reported, the posted record does not establish how much average change was observed or how variable participant outcomes were. It also does not report a confidence interval or statistical test result. Without these quantities, readers cannot judge the estimate’s numerical precision or determine how compatible the data are with a particular statistical hypothesis. The study’s enrollment figure does not fill that gap.

The single-group design also limits what a change over time could mean. A within-group change can describe the participants’ course during the study, but it does not provide a randomized estimate of what would have happened under an alternative. Other influences on change cannot be separated from the program using a control comparison that is not part of the recorded design. Therefore, the record cannot establish a comparative effect against another intervention or usual care.

The safety entries are similarly limited. They report no serious adverse events and no deaths, alongside 3 other adverse events, in the recorded group. These counts are not a comparative safety analysis, and the record does not provide detailed event descriptions. They should not be generalized beyond the group and reporting information recorded.

These limitations do not make the study question unimportant. They define what can be concluded from this registry record. The design and endpoint can be described, and the recorded counts can be reported, but claims about the magnitude, precision, or comparative effect of the intervention are not supported by the results recorded.

14. Why This Trial Matters Statistically

This study is statistically informative as an example of the distinction between a clearly named endpoint and a reported result. The registry specifies mean percent body weight loss at 4 months and identifies the measure as continuous, with a mean and standard deviation. Yet the numerical endpoint result is not reported in the registry record. Without the estimate and a measure of uncertainty, readers cannot evaluate the observed magnitude or precision, even though the endpoint is defined.

The single-group design also makes the target of inference important. A pilot can describe outcomes among participants exposed to a program and can inform questions for future research. But a before-and-after description is not the same as a randomized comparison. The design has no separate comparator, so it cannot directly estimate a contrast with an alternative condition. This distinction prevents readers from interpreting within-group change as a causal comparison.

The record’s planning illustration further highlights why design assumptions need to be read in context. It describes a two-arm planning framework, while the trial itself is recorded as single-group. Such an illustration is not the trial’s own design or an observed power calculation. Planning scenarios depend on assumptions and do not replace a reported result or an analysis plan.

Finally, the available safety counts show why descriptive reporting and comparative inference should remain separate. Counts can convey what was recorded in the study group, but without a comparator and event details they do not establish relative safety. Taken together, the record emphasizes the value of reporting estimates, uncertainty, analysis populations, missing-data methods, and design context alongside endpoint definitions.

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 trial-specific details needed for a fuller assessment are not reported in the registry record. Eligibility criteria, participant characteristics, recruitment methods, and the study setting are not described. The record therefore does not allow assessment of how participants were selected or how well the enrolled group represents the wider population of women with PCOS.

The intervention is described as a diet and lifestyle program and as an adapted very-low carbohydrate diet pilot, but implementation details are not reported. The record does not describe the program’s components, duration of participation, support or monitoring, adherence, or co-interventions. These details would help readers understand what exposure the intervention label represents and whether participants received it as intended.

Although enrollment is listed as 29, the number with primary endpoint data is not reported. The record does not explain missing follow-up measurements, the analysis population, or how incomplete data were handled. It also does not report the primary outcome estimate, its standard deviation, a confidence interval, or a statistical test. Thus neither the primary finding nor its precision can be assessed.

The design is single-group, with no separate comparator listed. This limits causal interpretation and prevents a direct estimate against another treatment or usual care. Masking is listed as none, and the record does not report further details about outcome measurement procedures that might help assess potential measurement influences.

Other analysis details, including the statistical model, covariate adjustment, multiplicity handling, and any changes from a planned analysis, are not reported in the registry record. The listed safety counts lack event descriptions and comparative context. These limits apply to interpretation of this registry record; they do not establish what may have been documented elsewhere.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: The University of Michigan PCOS Intervention Using Nutritional Ketosis — NCT03987854.
• Registry record last updated 2022-10-25; status last verified 2022-10.
• Results first posted 2022-10-25.
• 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.

19. Record Summary

The University of Michigan PCOS Intervention Using Nutritional Ketosis is recorded as a completed Phase 1 and Phase 2, single-group study with enrollment of 29. It began on 2019-07-05, with primary completion on 2020-03-01. Allocation is listed as not applicable, and masking as none.

The intervention is a behavioral diet and lifestyle program, described as a pilot test of an existing very-low carbohydrate diet intervention adapted for women with PCOS. The primary endpoint is mean percent body weight loss at 4 months, defined using the change from baseline weight. HbA1c change is listed as a secondary outcome.

The primary endpoint result, its variability, and its precision are not reported in the registry record. Safety entries list 0 serious adverse events, 3 other adverse events, and 0 deaths. The study has no listed comparator, so the design does not provide a randomized comparison.