Effects of Dapagliflozin on Insulin Resistance and Insulin Secretion in Subjects With Type 2 Diabetes: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Effects of Dapagliflozin on Insulin Resistance and Insulin Secretion in Subjects With Type 2 Diabetes (NCT00831779), a randomized phase 2 study. The primary outcome is Adjusted Mean Percent Change From Baseline in Insulin Sensitivity at Week 12 (Last Observation Carried Forward [LOCF]).

2. Clinical Question

The registry record describes a randomized, double-blind, placebo-controlled, parallel-group Phase 2 trial. Its stated purpose was to evaluate dapagliflozin’s effects on insulin sensitivity. The posted primary outcome operationalizes that question as an adjusted mean percent change from baseline in insulin sensitivity at Week 12, using last observation carried forward.

The record also reports a secondary outcome on insulin secretion. These measures address related but distinct aspects of insulin function. The posted results therefore give a statistical comparison for insulin sensitivity and a further comparison for insulin secretion. They do not, by themselves, establish how changes in either measure translate into clinical outcomes or longer-term effects.

3. Trial Design

This was a randomized, double-blind, placebo-controlled, parallel-group Phase 2 trial sponsored by AstraZeneca. The registry lists quadruple masking, but does not report which specific groups of people were masked. Participants were assigned to dapagliflozin 5 mg plus background anti-diabetes medication or placebo plus background anti-diabetes medication. Details of the background medication regimens are not reported in the registry record.

The registry lists enrollment of 116 and describes the study as completed. The recorded start was 2009-04, and primary completion was 2010-08. Those are registry dates; the record does not provide further detail about the timing of individual participant visits beyond the outcome timeframe. For the primary outcome, measurements were obtained at the randomization visit and Week 12 during the double-blind period.

The primary analysis population was defined as randomized participants who received study medication and had nonmissing values at baseline and Week 12, with the last observation carried forward. The record reports group-specific analysis counts for the posted outcomes; these counts are not the same as the enrollment figure. The original allocation ratio, the details of the randomization procedure, and any additional eligibility or treatment-adherence criteria are not reported in the registry record.

The registry identifies two outcome measures: the primary insulin-sensitivity measure and a secondary insulin-secretion measure. It reports an analysis for each. The trial design establishes a randomized comparison, but the record does not provide a complete statistical analysis plan or a sample-size justification for the original trial.

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: All randomized participants who received study medication and had nonmissing values at baseline and Week 12 (LOCF)

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 randomized participants who received at least 1 dose of study medication

6. Primary Endpoint

The registry lists the primary outcome as Adjusted Mean Percent Change From Baseline in Insulin Sensitivity at Week 12 (Last Observation Carried Forward [LOCF]), measured over From Baseline to Week 12.

The primary endpoint is the adjusted mean percent change from baseline in insulin sensitivity at Week 12. It expresses change relative to baseline as a percentage, rather than reporting only the Week 12 value. The registry defines baseline as the last assessment before the first dose of double-blind study medication. If dose or assessment time was unavailable, baseline was the last assessment on or before the date of the first dose.

Insulin sensitivity is the measure named in the trial’s primary question and stated purpose. Comparing its change between the randomized groups allows the analysis to assess whether the recorded pattern differed between dapagliflozin plus background medication and placebo plus background medication over the study period. The endpoint is reported as a mean, with standard error as the dispersion measure.

The primary timeframe runs from baseline to Week 12. The analysis used last observation carried forward (LOCF) for the Week 12 outcome. In general, LOCF substitutes a participant’s last available measurement for a later missing measurement. This can retain participants with incomplete follow-up, but it relies on an assumption about how the unobserved value relates to the last observed value. The registry record does not report other missing-data analyses that would show how sensitive the result is to that assumption.

The method used to measure insulin sensitivity, the clinical interpretation of a particular percentage change, and the threshold for a meaningful change are not reported in the registry record. The endpoint therefore supports an interpretation of the recorded statistical comparison, not a conclusion about clinical importance on its own.

7. Results Posted in the Registry

• Dapagliflozin 5 mg + Background Anti-Diabetes Medication: mean 7.98 versus -9.99 on control; difference 17.97 (95% CI 6.14 to 29.80), standardized effect d = 0.92; z-test p = 0.003.

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.

Adjusted Mean Change From Baseline in Insulin Secretion at Week 12 (Last Observation Carried Forward [LOCF])

• 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:
  • Dapagliflozin 5 mg + Background Anti-Diabetes Medication: mean 15.39 versus -12.73 on control; difference 28.12 (95% CI -0.06 to 56.30), standardized effect d = 0.61; z-test p = 0.051.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Dapagliflozin 5 mg + Background Anti-Diabetes Medication and the control. The interval gives the range of effects that remain compatible with the data.

The registry posts one secondary outcome: adjusted mean change from baseline in insulin secretion at Week 12. Its group means show an increase in the dapagliflozin group and a decrease in the placebo group, so the direction of the estimated contrast matches the direction of the primary group pattern. This offers a related measure of insulin function, but it is not the same endpoint as insulin sensitivity.

The secondary result is less certain statistically than the primary result. The recalculated interval includes zero, as does the interval from the sponsor’s reported ANOVA. This means the data are compatible with a range of contrasts that includes no difference under those analyses. It does not prove that the groups had identical outcomes, nor does a directional point estimate establish a reliable difference.

Secondary outcomes generally carry less inferential weight than a prespecified primary outcome, particularly when multiple outcomes or analyses are considered. Testing several outcomes can increase the chance of observing an apparently notable result by chance. The registry does not report a multiplicity adjustment or enough of the statistical plan to determine how secondary testing was intended to be interpreted. The secondary finding should therefore be viewed as supportive or exploratory evidence, rather than treated as an independent confirmation of the primary result.

The overall pattern is directionally consistent across the two posted measures, but the strength and precision of the statistical evidence are not the same. That pattern can motivate further study; it does not establish that changes in insulin secretion explain the primary outcome or that either measure predicts clinical outcomes.

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

• Adjusted Mean Change From Baseline in Insulin Secretion at Week 12 (Last Observation Carried Forward [LOCF])

9. Statistical Interpretation

10. Statistical Methods Explained

What does a mean difference represent?

For a continuous outcome, a group mean describes the average recorded value in that group. The mean difference compares the two group means, with the direction depending on which group is subtracted from which. For the primary outcome, the recalculated contrast is expressed in percentage points of change from baseline. It is not a relative percentage increase between the groups.

The arm-level results include a mean and a standard error. A standard error describes the statistical uncertainty in an estimated mean; it is not the spread of individual participant measurements. The registry does not post standard deviations for these outcomes. The recalculation uses the posted means, standard errors, and group counts to estimate an unadjusted contrast and its uncertainty. It is a summary-statistic calculation, not a reconstruction of all details of the trial’s analysis.

What does the confidence interval tell the reader?

A confidence interval gives a range of values reasonably compatible with the data under the statistical method and its assumptions. A narrower interval indicates greater precision, while a wider interval indicates more uncertainty about the size of the contrast. An interval that includes zero is compatible with no mean difference under this calculation; an interval that excludes zero is less compatible with that null value. This does not make a confidence interval the probability that the true effect lies within its particular endpoints.

For the primary recalculation, the interval lies above zero. The sponsor’s reported interval also lies above zero. For the secondary outcome, both the recalculated and sponsor-reported intervals extend across zero. These are statistical descriptions of uncertainty, not stand-alone tests of clinical relevance.

Why are the recalculation and ANOVA estimates different?

The recalculated comparisons are simple, unadjusted estimates based on the posted group summary statistics. The registry identifies the sponsor’s analyses as ANOVA and labels their parameter as a mean difference in final values. An analysis of final values and an unadjusted comparison of posted means need not give identical estimates or intervals. The registry record does not provide the complete ANOVA specification, covariates, or model diagnostics, so the sponsor’s analysis cannot be independently reconstructed in full from the record.

ANOVA is a general method for comparing means across groups. In a randomized trial it can be used to compare outcome values, and a model may include additional terms if its specification calls for them. Which terms were used matters for interpreting an adjusted estimate. The fact that the registry labels the method ANOVA does not reveal all those details.

What role does last observation carried forward play?

LOCF uses the last available measurement as a substitute for an unobserved later measurement. It is a missing-data convention, not a measurement of the participant’s actual Week 12 value. Its suitability depends on how outcomes change over time and why observations are missing. If outcomes would have changed after the last measurement, carrying the value forward can misrepresent the later value. The registry defines the primary population using nonmissing baseline and Week 12 values with LOCF, but does not report the amount or reasons for missing data, or sensitivity analyses using alternative assumptions.

Why can over-dispersion matter in other analyses?

General statistical context: over-dispersion occurs when observed variability exceeds what a model assumes. It can make standard errors too small if the model does not account for the extra variability, which may make confidence intervals too narrow and p-values too small. The posted outcomes in this trial are continuous means, and the registry record does not report an over-dispersion assessment. This concept should not be taken as a finding about this trial.

What can and cannot be inferred from the p-values?

A p-value describes how unusual the observed data, or more extreme data, would be under a specified null model and its assumptions. It does not give the probability that the null hypothesis is true, and it does not measure effect size or clinical importance. The recalculated p-values use unadjusted summary-statistic comparisons; the sponsor’s p-values come from the reported ANOVA analyses. Because the record does not report a full analysis plan, model specification, or multiplicity procedure, the p-values should be read in that context rather than as a complete account of all planned inference.

11. Confidence Intervals and Precision

For the primary endpoint, the recalculated mean difference is 17.97 percentage points, with a 95% confidence interval from 6.14 to 29.8. The interval is the key indication of precision: it gives a range of contrasts compatible with the unadjusted calculation, rather than a single certain value. It excludes zero, but still spans effects of different sizes. The point estimate should therefore not be read as a perfectly known effect.

The sponsor’s reported primary ANOVA result is a mean difference of 19.97, with a 95% confidence interval from 5.75 to 36.10. Its interval also excludes zero, but has different endpoints from the recalculated interval. The estimates are not identical because the reported ANOVA analysis of final values and the summary-statistic recalculation are different analyses. The registry does not report enough model detail to determine precisely which features account for the difference.

For the secondary outcome, the recalculated confidence interval extends from -0.062 to 56.3, while the sponsor’s interval extends from -1.22 to 57.48. Both include zero. The breadth of these intervals indicates uncertainty about the size and direction of the mean contrast, even though the posted group means have different values.

Confidence intervals depend on the statistical model and its assumptions. They do not account automatically for every possible source of bias, missing-data uncertainty, or multiple testing. In this trial, the primary analysis uses LOCF and the registry does not report alternative missing-data analyses or a multiplicity procedure. Precision should thus be understood as conditional on the stated analysis and available record, not as a measure of every uncertainty relevant to interpretation.

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, not the number of participants actually enrolled or analyzed. For these continuous outcomes, the figures are participants per arm, not total event counts. The illustrations assume a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. They use the observed effect as their starting point and show planning for the full observed effect, 75% of the observed effect, and 50% of the observed effect.

The 80% power column means the illustrative number of participants per arm associated with an 80% chance of detecting the stated effect under those planning assumptions, if the effect and assumptions were correct. The 90% power column expresses the corresponding planning target for a 90% chance. These are design calculations, not evidence about the probability that the completed trial detected an effect.

For the primary comparison, planning for 100% of the observed effect gives 19 participants per arm for 80% power and 25 participants per arm for 90% power. Planning for 75% gives 33 and 45 participants per arm, respectively; planning for 50% gives 75 and 100. For the secondary comparison, the corresponding pairs are 43 and 58 for the full observed effect, 76 and 102 for 75%, and 171 and 229 for 50%.

Smaller effects require more participants per arm to reach the same power target under the stated assumptions. These values are illustrations computed from observed estimates, not the trial’s original design calculations; the original sample-size assumptions and justification are not reported in the registry record. This is not observed power.

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 = 17.970; pooled standard deviation σ = 19.528.
2. n = 2 × (1.96 + 0.842)² × 19.528² / 17.970² = 18.5, rounded up to 19 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 <- 17.97                      # observed difference in means
sigma <- 19.5281494859639                     # 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(831779)
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 = 19 (80% power), 25 (90% power)
 75% of observed effect: n per arm = 33 (80% power), 45 (90% power)
 50% of observed effect: n per arm = 75 (80% power), 100 (90% power)
power.t.test: 19.5 
Simulated power at n = 19 per arm: 0.788 (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

The registry reports the sponsor’s own analysis for each outcome as an ANOVA comparison of final values. For the primary outcome, that analysis reports a mean difference of 19.97, a 95% confidence interval from 5.75 to 36.10, and p = 0.006. The unadjusted recalculation from posted summary statistics reports 17.97, a 95% confidence interval from 6.14 to 29.8, and p = 0.003.

These estimates should be kept distinct. The recalculation is a simple comparison of posted group summary statistics; the sponsor’s analysis is identified as ANOVA and concerns final values. The registry record does not provide a complete model specification that would allow an exact reproduction of the sponsor’s analysis or a detailed explanation of the numerical differences. Both reported primary intervals exclude zero, but they are not interchangeable estimates.

The registry also reports an ANOVA result for insulin secretion, with a mean difference of 28.13, a 95% confidence interval from -1.22 to 57.48, and p = 0.060. Its interval includes zero. The recalculated secondary contrast is a separate unadjusted result. The registry does not indicate a discrepancy flag, but the sponsor’s reported analyses remain the trial’s own analyses and should be distinguished from summary-statistic recalculations.

15. Multiplicity, Missing Data and Analysis Populations

The registry identifies a primary outcome and a secondary outcome. A primary outcome usually anchors the main statistical question, while secondary outcomes address additional questions. When several outcomes, timepoints, subgroups, or analyses are tested, the chance of at least one apparently unusual result can increase. Statistical procedures that account for multiple testing can help control this issue, but the registry record does not report a multiplicity adjustment or a detailed testing hierarchy.

The primary result is therefore the central posted comparison, while the secondary result warrants more caution. The absence of a reported adjustment does not prove that none was used in the full study process; it means the procedure is not reported in the registry record. The record also does not specify whether other outcomes or exploratory analyses were conducted but not posted.

The primary analysis population is described as randomized participants who received study medication and had nonmissing baseline and Week 12 values, with LOCF used for the Week 12 endpoint. This is not an analysis of every randomized participant as described by an intention-to-treat principle. The registry does not report the number or characteristics of participants excluded from the primary population, the amount of missing data, or the reasons for missingness. Group-specific analysis counts appear with the outcome results, but the record does not provide a participant flow account sufficient to explain all differences in counts across outcomes.

LOCF addresses missing Week 12 observations by carrying forward the last available measurement. As a general statistical matter, this approach assumes that the carried-forward value is a reasonable substitute for the missing value. That assumption may not hold if the outcome changes after the last observation. The registry does not report sensitivity analyses using alternative missing-data methods, so the robustness of the results to that choice cannot be assessed from this record.

The original statistical analysis plan, including any planned interim analyses, is not reported in the registry record. No conclusion should be drawn about whether interim analyses or other multiplicity controls were planned or conducted.

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 5% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports zero serious adverse events in the placebo group among 21 participants at risk and zero in the dapagliflozin group among 23 participants at risk. These are the serious adverse event figures posted for the two groups; they should be read with their stated at-risk counts and the limitations of the recorded safety information.

The registry also lists other adverse events affecting 8 participants in the placebo group and 9 in the dapagliflozin group, with 21 and 23 participants at risk, respectively. These are reported separately from serious adverse events. They do not describe event severity, timing, duration, or causal attribution in detail.

Deaths are not reported in the registry record. The record does not list serious adverse event terms. These trial counts do not establish the absence of risk, and they do not support a broad safety comparison beyond the specific recorded figures. Interpretation is also limited by the trial’s recorded timeframe and the safety details available in the registry.

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

The primary comparison shows that the estimated mean percent change in insulin sensitivity was higher in the dapagliflozin group than in the placebo group at Week 12. Both the unadjusted recalculation and the sponsor’s reported ANOVA result have confidence intervals that exclude zero. This is evidence of a between-group difference for the posted primary measure under those analyses.

That statistical result is not, by itself, a demonstration of clinical benefit. The registry record does not report a threshold for a clinically meaningful change in insulin sensitivity, nor does it connect the measured change to symptoms, complications, or other patient-important outcomes. The endpoint is a laboratory or physiological measure as named in the record, and its clinical implications cannot be determined from the posted comparison alone.

The secondary insulin-secretion comparison is directionally consistent in the group means, but its intervals include zero. It consequently provides less precise statistical evidence for a difference. Neither the primary nor secondary result explains a biological mechanism, and the registry does not establish that a change in one measure caused a change in the other.

These results concern the recorded groups, background medication context, analysis populations, and Week 12 timeframe. They do not establish longer-term effects, effects in populations not represented by the trial, or outcomes beyond the measures posted in the registry. The results should not be interpreted as medical advice or as a treatment recommendation.

Finally, statistical significance and clinical importance are separate questions. A confidence interval and p-value describe uncertainty under a particular analysis; they do not account for all potential limitations, including missing-data assumptions, unreported analysis details, or the interpretation of the endpoint in clinical practice.

18. Why This Trial Matters Statistically

This trial illustrates why a registry result should be read through both its endpoint definition and its analysis method. The primary measure is a percent change from baseline, analyzed at Week 12 with LOCF, while the sponsor’s reported analysis is an ANOVA comparison of final values. A simple recalculation from group summaries is useful for understanding the approximate contrast and its uncertainty, but it is not a substitute for the trial’s full analysis.

The primary estimate is accompanied by a confidence interval, making it possible to consider both the direction and precision of the comparison rather than relying only on a p-value. The secondary result also demonstrates why an estimate’s direction should not be mistaken for firm evidence: its interval includes zero despite the directional group means.

The trial further highlights the distinction between a planning illustration and an original trial design calculation. Sample-size illustrations based on an observed effect can show how the required number of participants per arm changes when the target effect is smaller. They cannot establish the design assumptions used for the completed study or describe its achieved power.

Finally, the record’s limited reporting of multiplicity, missingness, and model details matters. Statistical conclusions depend not only on the observed estimates but also on how outcomes were selected, how missing data were handled, and what model generated the reported inference. A transparent reading separates those the registry entries from details that are not reported.

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 uses LOCF, but the registry does not report how much data were missing, why observations were missing, or whether conclusions were checked under alternative missing-data assumptions. The stated analysis population also requires receipt of study medication and nonmissing baseline and Week 12 values, so it is not described as including all randomized participants regardless of treatment receipt or endpoint availability.

The sponsor’s reported analysis is identified as ANOVA on final values, but the registry does not report the complete model specification, covariates, diagnostics, or handling of any baseline differences. This limits independent reconstruction and makes it difficult to explain fully why the sponsor’s estimates differ from the unadjusted recalculations based on posted summaries.

The registry record does not report a multiplicity procedure, a detailed statistical analysis plan, the original sample-size justification, or whether interim analyses were planned or conducted. It also does not provide a complete account of participant flow or explain the differing analysis counts across the posted outcomes. Details of the insulin measurement methods and a threshold for clinical importance are not reported.

The outcome timeframe is baseline to Week 12. The record does not provide evidence about longer-term changes or clinical outcomes beyond insulin sensitivity and insulin secretion. The safety information is limited to posted adverse-event counts; deaths are not reported in the registry record, and the available entries do not provide a full clinical characterization of events.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Effects of Dapagliflozin on Insulin Resistance and Insulin Secretion in Subjects With Type 2 Diabetes — NCT00831779.
• Registry record last updated 2017-04-24; status last verified 2017-03.
• Results first posted 2017-03-01.
• 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

The registry record describes a completed, randomized, double-blind, placebo-controlled, parallel-group Phase 2 trial sponsored by AstraZeneca. It lists enrollment of 116 and compares dapagliflozin 5 mg plus background anti-diabetes medication with placebo plus background anti-diabetes medication in subjects with type 2 diabetes mellitus.

The primary outcome is adjusted mean percent change from baseline in insulin sensitivity at Week 12, analyzed using LOCF in the stated population. The posted primary estimates favor a higher mean change in the dapagliflozin group, and the recalculated and sponsor-reported confidence intervals exclude zero. The secondary outcome concerns insulin secretion; its reported intervals include zero.

The registry provides limited information about missing data, multiplicity, the full ANOVA specification, and the trial’s original sample-size rationale. The results describe the recorded statistical comparisons and should not be extended beyond the reported measures and timeframe.