LOSS- Louisiana Obese Subjects Study: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for LOSS- Louisiana Obese Subjects Study (NCT00115063), a randomized phase 4 study. The primary outcome is Percent Change From Baseline Weight.

2. Clinical Question

3. Trial Design

LOSS (Louisiana Obese Subjects Study) was a phase 4, randomized, parallel-group trial sponsored by Pennington Biomedical Research Center. The registry lists an enrollment of 390 and a start date of 2005-07. The status is terminated, and the primary completion date is 2007-12. The reason for termination is not reported in the registry record.

The two arms were an intensive medical intervention and access to a weight-loss informational website. The brief summary describes the intensive approach as aiming to produce 25% weight loss from baseline and maintain at least 20% weight loss from baseline. These are stated treatment aims, not the observed primary result. The same summary describes an active treatment period of three years followed by two years of observation. The posted primary outcome, however, is measured at the two-year visit.

Masking is listed as none. The registry does not report additional details about allocation concealment, the randomization procedure, treatment adherence, or how participants were assigned to the listed medications. The primary analysis population consisted of completers of the two-year study visit, with 101 participants in the intensive medical intervention group and 86 in the website-access group. The record does not explain how participants who did not complete that visit were handled in analyses.

The listed outcomes include weight, blood pressure, blood tests, fasting plasma glucose, and the Duke Activity Status Index questionnaire. Results are available for the primary percent-change measure and several secondary measures. Other listed outcomes are not reported in the registry record in the detailed results.

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: This analysis was done on the completers, those that completed the 2 year study visit.

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 Percent Change From Baseline Weight, measured over Baseline, 2 years.

The primary endpoint is mean percent change in weight from baseline to two years. Expressing change as a percentage relates the change to each participant’s starting weight, while reporting a group mean describes the average change among participants included in the analysis. Negative values indicate a decrease from baseline; values closer to zero indicate less average change in that direction.

The endpoint is useful for comparing the average weight trajectory of the randomized groups over a specified period. The intensive medical intervention group’s posted mean was -9.7%, and the website-access group’s mean was -0.4%. The estimated mean difference was 9.3 percentage points when expressed as website-access minus intensive-intervention change. The positive contrast therefore reflects the more negative mean change in the intensive intervention group.

This is a continuous outcome, and the reported arm-level dispersion is the standard error of each mean. The primary analysis uses completers of the two-year visit, rather than all participants enrolled. It therefore describes the analyzed completer groups; the registry record does not report an analysis that includes missing two-year outcomes or establishes how the completers compare with those who did not complete the visit.

7. Results Posted in the Registry

• Access to Weight Loss Informational Website: mean -0.40 versus -9.70 on control; difference 9.30 (95% CI 6.41 to 12.19), standardized effect d = 0.88; z-test p = <0.001.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Change in Weight From Baseline in Kilograms (kg)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Access to Weight Loss Informational Website: mean -0.50 versus -12.70 on control; difference 12.20 (95% CI 8.43 to 15.97), standardized effect d = 0.89; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically higher for Access to Weight Loss Informational Website than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Change in Fasting Plasma Glucose (FPG) in Milligrams Per Deciliter (mg/dL)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Access to Weight Loss Informational Website: mean 6.70 versus 2.40 on control; difference 4.30 (95% CI -1.52 to 10.12), standardized effect d = 0.21; z-test p = 0.148.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Access to Weight Loss Informational Website and the control. The interval gives the range of effects that remain compatible with the data.

Change in Duke Activity Status Index (DASI) Questionnaire Score

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard errors.
• Result:
  • Access to Weight Loss Informational Website: mean -11.60 versus -2.10 on control; difference -9.50 (95% CI -14.64 to -4.36), standardized effect d = -0.53; z-test p = <0.001.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Access to Weight Loss Informational Website than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcome results add several perspectives to the primary weight endpoint. The posted results include weight change in kilograms, fasting plasma glucose, and Duke Activity Status Index score. The overall pattern is not uniform across measures: the weight measures describe greater average reduction in the intensive-intervention group, while the glucose result does not show a clear separation in its reported interval and test. The activity-status measure also has a between-group contrast, but its direction is tied to the scale’s score change and should not be treated as interchangeable with weight change.

These outcomes relate to the primary result by describing other measured changes over the same baseline-to-two-year period. They do not independently explain the primary weight contrast or identify which part of the combined intervention may be related to any observed difference. The registry lists additional outcomes concerning blood pressure and blood tests, but their detailed results are not reported in the registry record.

The secondary findings carry less weight for confirmatory interpretation because the registry does not report a multiplicity adjustment or a hierarchy for interpreting the multiple outcomes. In general, testing multiple outcomes increases the chance that at least one result will appear noteworthy by chance if each test is treated as independent evidence. The registry reports ten analyses, but does not state how the analysis set was prespecified or how multiplicity was handled. Accordingly, the secondary results are best read as additional descriptive evidence rather than as a collection of independent confirmations of the primary finding.

The primary endpoint remains the central reported comparison. Differences in measurement scales, missing detailed results for some listed outcomes, and lack of a stated multiplicity strategy limit how broadly the secondary pattern can be interpreted.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Change in Weight From Baseline in Kilograms (kg)
• Change in Blood Pressure
• Percent Change in Blood Tests- Low Density Lipoprotein (LDL), High Density Lipoprotein (HDL), Triglycerides and Uric Acid
• Change in Fasting Plasma Glucose (FPG) in Milligrams Per Deciliter (mg/dL)
• Change in Duke Activity Status Index (DASI) Questionnaire Score

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: each analyzed participant contributes a percent change in weight, and the reported arm result is a mean. The registry gives standard errors for the group means rather than standard deviations. A standard error describes the precision of an estimated mean; it is not a measure of the spread of individual participant values. The arm means and standard errors are -9.7% and 1.3 for the intensive intervention, and -0.4% and 0.7 for website access.

What does the reported test assess?

The registry reports a two-sided t-test comparing the two groups. In general, a two-sided test evaluates evidence for a difference in either direction, rather than prespecifying that only one direction can count as a difference. A t-test compares group means while accounting for variability and sample size. The posted P value is < 0.001. A P value describes how incompatible the observed data are with the test’s null model of no group-mean difference, under the test assumptions. It is not the probability that the null hypothesis is true, nor a measure of the size or practical importance of a result.

How is the contrast expressed?

The reported comparison is expressed as website-access mean minus intensive-intervention mean. Because the arm means are negative, the estimated difference is positive: 9.3 percentage points. The same comparison can be understood by noting that the intensive-intervention group had the more negative average percent change. The recalculated confidence interval is unadjusted and uses the posted arm summaries; it is not an adjusted estimate from a model described by the registry.

What is a confidence interval?

In general, a confidence interval gives a range of values compatible with the data and the statistical procedure used, at a stated confidence level. The posted recalculated interval is 95%, from 6.406 to 12.19 percentage points. Its width reflects uncertainty in the mean contrast. It does not mean that 95% of individual outcomes fall within that range, and it does not account for possible bias from missing outcomes or other design limitations.

What does randomization contribute?

In general, random assignment is intended to make groups comparable on average at the beginning of a study, including on factors that may not be measured. It supports interpretation of between-group differences as associated with assigned group, subject to issues such as follow-up, adherence, and analysis choices. The record lists the allocation as randomized but does not report the randomization procedure, allocation concealment, or a full analysis plan. The primary analysis’s restriction to visit completers is important: if completion differs by group or relates to outcomes, the original balance from randomization may not be retained among completers.

Why can over-dispersion matter?

In general, over-dispersion occurs when outcome variation is greater than a statistical model assumes. If a model understates variation, its standard errors can be too small and its intervals too narrow. For a continuous-outcome comparison, appropriate uncertainty estimates depend on the actual variability and the assumptions used. The registry reports a t-test and standard errors, but it does not report diagnostics for distributional assumptions, a model for over-dispersion, or an alternative robust analysis. The registry record therefore does not establish whether such checks were performed.

What is recalculated?

The displayed recalculated contrast and interval are simple, unadjusted calculations from the posted summary statistics, rather than a substitute for the sponsor’s analysis. The sponsor’s reported test is also a two-sided t-test and gives p < 0.001 for the primary outcome. The registry does not report the sponsor’s test statistic, exact P value, or confidence limits for that outcome. No discrepancy is flagged between the reported and recalculated results.

In general, statistical significance and effect magnitude answer different questions. The P value concerns compatibility with a null model, while the estimated difference and its interval convey the scale and precision of the mean contrast. Neither alone resolves the influence of incomplete follow-up, the relevance of the endpoint to participants, or the effects of considering multiple outcomes.

11. Confidence Intervals and Precision

The primary mean difference is 9.3 percentage points, and its recalculated unadjusted 95% confidence interval runs from 6.406 to 12.19. In practical statistical terms, the interval gives a range of plausible values for the group mean contrast under the calculation and assumptions used. Its limits are both positive, consistent with the estimated direction in which the website-access group had the less negative mean percentage change.

Confidence intervals express uncertainty about a group-level estimate, not variability among individual participants. The arm standard errors are 1.3 for the intensive medical intervention mean and 0.7 for the website-access mean. They describe precision of the reported means, not the standard deviation of individual weight changes. The recalculated comparison uses the posted summary statistics, and the interval is not an adjusted interval from a reported covariate model.

The interval does not address uncertainty arising from unreported analysis choices, missing two-year outcomes, or the completer-only population. It also does not imply that the true difference has a 95% probability of lying inside this particular interval. In general, the 95% label refers to the long-run coverage of the procedure under its assumptions. Since the registry reports no missing-data analysis or sensitivity analysis for the primary endpoint, those sources of uncertainty are not reflected in the posted interval.

The sponsor reports p < 0.001 but does not report primary confidence limits or the exact test statistic in the registry record. The interval therefore gives a useful precision description from the posted arm summaries, while the sponsor’s own reported analysis remains the registry’s inferential result.

12. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The sample-size figures are planning illustrations based on the observed effect, not the trial’s original design calculations. They assume a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The assumed effect to detect is the stated fraction of the observed effect. The original trial’s sample-size justification and design assumptions are not reported in the registry record.

For the primary endpoint, the figures are participants per arm. They are not total events: the event-count convention applies to time-to-event outcomes, whereas this is a continuous measure. The planning display gives separate estimates for 80% and 90% power. In plain language, the 80% column estimates how many participants in each arm would be needed for an 80% chance of detecting the assumed difference under the stated planning assumptions if that difference were the true effect. The 90% column expresses the corresponding planning target for a 90% chance.

The rows consider planning for 100%, 75%, and 50% of the observed effect. As the assumed difference gets smaller, more participants are needed to detect it at the same power target. This illustrates why planning directly from an observed effect can be optimistic when that estimate is uncertain, and why considering smaller effects can change the required sample size. These figures are not observed power: they are hypothetical sample-size calculations, not a retrospective measure of the probability that the completed trial detected its actual effect.

The displayed calculations concern a simplified two-group comparison. They do not establish what sample size would have been appropriate under other design assumptions, an adjusted analysis, or a different approach to incomplete outcome data. No such original design details are reported in the registry record.

13. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = 9.300; pooled standard deviation σ = 10.565.
2. n = 2 × (1.96 + 0.842)² × 10.565² / 9.300² = 20.3, rounded up to 21 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- 9.3                      # observed difference in means
sigma <- 10.5653436520771                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(115063)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 21 (80% power), 28 (90% power)
 75% of observed effect: n per arm = 37 (80% power), 49 (90% power)
 50% of observed effect: n per arm = 82 (80% power), 109 (90% power)
power.t.test: 21.3 
Simulated power at n = 21 per arm: 0.798 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

14. Analyses Reported by the Sponsor

For the primary outcome, the registry reports that the sponsor used a two-sided t-test and gives p < 0.001. It does not report the sponsor’s numerical estimate of the between-group difference, confidence limits, test statistic, or exact P value. The posted arm means and standard errors allow an unadjusted recalculation of the difference and its 95% confidence interval: 9.3 percentage points, with limits from 6.406 to 12.19. The recalculated P value is < 0.001.

The registry does not flag a discrepancy between the sponsor’s reported result and the recalculation. The recalculation is based on summary statistics and is not a reproduction of any individual-level or adjusted analysis. When a discrepancy is present, the sponsor’s own analysis takes precedence; no discrepancy is reported for this comparison. The sponsor’s analysis is the inferential result recorded in the registry, while the recalculated interval adds a simple unadjusted description of precision.

The sponsor also reports two-sided t-tests for the posted secondary outcomes. For fasting plasma glucose, the sponsor’s P value is 0.16, whereas the recalculated unadjusted P value is 0.148. For the Duke Activity Status Index, the sponsor reports 0.001 and the recalculated P value is < 0.001. The registry does not provide sponsor confidence limits for these outcomes. Differences in displayed P-value precision should not be treated as evidence of a substantive conflict.

15. Multiplicity, Missing Data and Analysis Populations

The registry lists six outcome measures and ten analyses. It does not report whether the outcomes and analyses were all prespecified, whether one endpoint was formally designated as the sole confirmatory test in the analysis plan, or whether any adjustment for multiple testing was used. The primary outcome is identified in the record, but the broader analysis plan and any procedure for controlling multiplicity are not reported in the registry record.

Multiplicity matters because the more outcome comparisons researchers examine, the greater the general opportunity for apparently unusual results to arise by chance. This does not invalidate a reported result, but it affects how strongly a set of unadjusted secondary P values can be interpreted. The registry reports t-tests for the primary endpoint and posted secondary results, without describing a family of tests or a correction. The secondary results should therefore be considered in light of that uncertainty.

The primary analysis was conducted among participants who completed the two-year study visit. The registry does not report the number of missing primary outcomes by randomized group, reasons for noncompletion, or a method for handling missing outcomes. It also does not report an intention-to-treat analysis, imputation, or sensitivity analyses that test assumptions about missing data. These details are important because a completer analysis can differ from an analysis of all randomized participants if completion is related to assignment or outcome.

The record does not describe exclusions after randomization, protocol deviations, or whether the analysis population was defined before results were examined. It also does not report interim analyses or stopping rules. These matters should be treated as not reported, rather than assumed absent. The terminated status is recorded, but the reason for termination is not reported in the registry record.

16. 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 4% reporting threshold.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: hernia repair (1.0%), chest pain (1.0%), total knee replacement surgery (0.5%), hospitalization for acute myocardial infarction (0.5%), pacemaker insertion procedure (0.5%).

The registry records serious adverse events affecting 20 participants in the intensive medical intervention group, with 200 at risk, and 8 participants in the website-access group, with 190 at risk. These are the registry’s serious-event figures; they should not be confused with the smaller numbers included in the primary completer analysis. The record does not explain the relationship between the safety denominators and the two-year outcome population.

The listed serious-event terms and percentages include hernia repair and chest pain at 1.0% each, and total knee replacement surgery, hospitalization for acute myocardial infarction, pacemaker insertion procedure, and hospitalization for back surgery at 0.5% each. These entries describe reported terms; the registry listing does not establish that a particular event was caused by an intervention.

The non-serious adverse-event fields list zero at risk and zero affected in both groups. Those entries do not provide an interpretable account of non-serious events, and should not be read as proof that none occurred. Death counts are not reported in the registry record. The available figures do not by themselves support a comparative safety conclusion.

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

The primary result describes a difference in average percent weight change among participants who completed the two-year visit. The intensive medical intervention group had a mean change of -9.7%, compared with -0.4% for website access. The estimated difference was 9.3 percentage points, with a recalculated unadjusted 95% confidence interval from 6.406 to 12.19. The registry reports p < 0.001 for its two-sided test.

This is a group-level result. It does not mean that every participant in one group changed in the same way, and it does not give the distribution of individual outcomes. The confidence interval describes precision around the estimated group contrast under the calculation used, not the range of likely individual changes. The completer-only analysis also means the result cannot automatically be generalized to every randomized participant, particularly because missing-outcome handling is not reported.

The intervention combined dietary, medication, and behavioral elements. The comparison does not isolate the contribution of any single element, and the registry does not report details that would support separating their effects. The result also does not establish long-term outcomes beyond the reported two-year endpoint. Although the brief summary describes a longer active treatment and observation plan, the primary result discussed is from the two-year visit.

The statistical evidence does not by itself determine the clinical importance of the observed contrast. It also does not establish effects on outcomes that are not reported, explain the reason for trial termination, or resolve the possible influence of multiple testing and missing outcomes. The findings are best understood as the posted comparison of two randomized groups on a defined measure and analysis population, together with the limitations of the registry record.

18. Why This Trial Matters Statistically

This trial illustrates the distinction between a randomized design and the population actually represented in a reported analysis. Random assignment is intended to support a fair group comparison, but the primary endpoint is reported for two-year completers. Without the number and characteristics of noncompleters or a missing-data analysis, readers cannot determine how closely that subset reflects the full randomized groups.

It also shows why an effect estimate, confidence interval, and P value answer different questions. The estimate gives the scale and direction of the average contrast; the interval conveys its statistical precision; and the P value summarizes evidence against a null model under the reported test. Reading these together is more informative than treating a small P value as a complete account of the result.

The intervention is a combination of components, so the observed between-group difference cannot identify which component contributed to the group-level result. The record also lists several secondary outcomes but does not describe multiplicity handling. This makes the trial a useful example of why the primary endpoint, secondary findings, analysis population, and statistical plan should be considered separately when interpreting clinical trial results.

Finally, the posted sample-size figures are planning illustrations derived from observed effects, not the original design justification. They demonstrate how planning targets depend on the effect one seeks to detect and on assumptions about power and allocation. They should not be mistaken for evidence about the trial’s design adequacy or its observed probability of success.

19. 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 primary analysis includes participants who completed the two-year visit, not necessarily all randomized participants. The registry does not report missing primary outcomes by group, reasons for noncompletion, or methods used to address missing data. This limits assessment of whether the completer comparison retained the balance intended by randomization.

The registry record does not report the original sample-size justification, detailed randomization or allocation-concealment procedures, an analysis plan, interim analyses, or stopping rules. The trial is listed as terminated, but the reason is not reported. The record also does not provide enough detail to assess medication selection, adherence, intervention fidelity, or which components of the intensive program participants received.

The sponsor’s primary report includes a two-sided t-test and P value but not its numerical estimate, confidence limits, or test statistic. The recalculated confidence interval is an unadjusted calculation from summary data, rather than an adjusted or individual-participant analysis. No alternative model or sensitivity analysis is reported.

Several outcomes are listed, while detailed results for some are not reported in the registry record. The record does not describe multiplicity control, and the detailed safety information is limited, including death counts that are not reported. Together, these omissions constrain interpretation of the breadth, robustness, and generalizability of the posted findings.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: LOSS- Louisiana Obese Subjects Study — NCT00115063.
• Registry record last updated 2016-02-22; status last verified 2016-01.
• Results first posted 2010-06-07.
• 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.

23. Record Summary

LOSS was a phase 4 randomized, parallel-group study sponsored by Pennington Biomedical Research Center, with an enrollment of 390. It compared an intensive medical intervention with access to a weight-loss informational website. The registry lists no masking and records the trial status as terminated.

The primary result concerns percent change from baseline weight at two years among completers. Mean changes were -9.7% and -0.4%, respectively; the estimated mean difference was 9.3 percentage points, with a recalculated unadjusted 95% confidence interval from 6.406 to 12.19. The sponsor reports a two-sided t-test with p < 0.001.

Interpretation is limited by the completer analysis, unreported missing-data handling, and absence of a reported multiplicity strategy. The registry does not report the reason for termination or the trial’s original sample-size justification.