Exercise and Inflammation in the Elderly: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Exercise and Inflammation in the Elderly (NCT00753363), a non-randomized na study. The primary outcome is Insulin Sensitivity.

2. Clinical Question

The study’s background describes persistent low-grade inflammation and impaired fibrinolysis as concerns in the older Veteran population, and identifies insulin resistance as a relevant mechanism. It also motivates the study by noting interest in whether lifestyle changes could have anti-inflammatory effects. Those statements describe the rationale for the research, not findings from this trial.

The central statistical question would be how the recorded insulin-sensitivity measure changed over time in each arm and how those changes compared. Answering it requires outcome values and an analysis that accounts for the study’s non-randomized allocation. Neither numerical endpoint results nor an analysis are reported in the registry record. The stated clinical question can therefore be described, but its answer cannot be inferred from the record.

3. Trial Design

The registry record describes a completed, parallel, non-randomized study sponsored by the VA Office of Research and Development. Its listed start date is 2007-04-01 and its primary completion date is 2012-11-29. Enrollment is reported as 119. The phase is listed as NA, and masking is listed as none.

Participants were assigned to one of two intervention arms: 6 months of aerobic exercise training or 6 months of weight loss. Because allocation was non-randomized, the arms should not be treated as equivalent groups created by random assignment. Differences between their outcomes, if results were available, could reflect differences in the participants as well as differences associated with the interventions. The registry record does not report how allocation decisions were made or whether any statistical adjustment was planned to address group differences.

The primary endpoint was insulin sensitivity, assessed at baseline and 6 month. Body weight and fitness are listed as secondary outcomes. The record reports 3 outcome measures, but it does not give the numerical results for the primary endpoint or the secondary outcomes. It also lists 0 analyses. Thus, the record describes the study design and the intended measurement framework without supplying an inferential comparison of the interventions.

The registry record does not report a protocol-level sample-size justification, an interim-analysis plan, or detailed procedures for assigning participants to arms. The information available supports a description of the study as a non-randomized, parallel comparison, but not a claim that the design can isolate causal effects of either intervention. The completed status indicates that the study is recorded as completed; it does not supply the missing outcome estimates.

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.

6. Primary Endpoint

The registry lists the primary outcome as Insulin Sensitivity, measured over Baseline and 6 month.

Insulin sensitivity is the primary endpoint. The registry record defines insulin resistance as a reduction in glucose utilization rate elicited by a given insulin concentration. Glucose utilization is measured during a three hour hyperinsulinemic-euglycemic clamp and presented as a measure of insulin sensitivity. The record describes the clamp as one of the most sophisticated methods for measuring insulin sensitivity.

The endpoint is reported as a continuous measure, with the mean as the stated summary and standard error as the stated measure of dispersion. Its unit is μmol/kgFFM/min. Measurements are specified at baseline and 6 month, so the planned measurement framework involves observations at both time points. The registry record does not provide the mean values, standard errors, changes from baseline, or an arm-to-arm estimate.

In general, an insulin-sensitivity measure is used to describe how glucose utilization responds under the specified insulin conditions. A mean gives a measure of the group’s average value, while a standard error describes the precision of an estimated mean. Neither the endpoint definition nor the measurement method provides evidence about the direction or size of a treatment effect. Without the endpoint results and their analysis, the registry record cannot show whether either arm changed over time or whether the arms differed.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record identifies a continuous primary outcome summarized by its mean and standard error, measured at baseline and 6 month. It also describes the study as parallel and non-randomized, with no masking. These design and measurement details frame how results would need to be interpreted, but the record does not include an outcome analysis or numerical endpoint values.

What does a mean describe?

In general, a mean is a summary of the values observed in a group. For a continuous endpoint such as insulin sensitivity, it can describe the group’s average at a specified measurement time. A mean alone does not describe how widely individual values vary, and it does not show whether a difference between groups is attributable to an intervention. The registry record’s endpoint description specifies means, but the means themselves are not reported.

How should the standard error be read?

In general, a standard error describes the sampling precision of an estimated statistic, such as a mean. It is not the same as a measure of the spread of individual participant values. Smaller standard errors generally indicate a more precisely estimated mean, while larger standard errors indicate greater uncertainty about that estimate. Here, no standard errors for the endpoint are reported, so the precision of the primary outcome estimates cannot be assessed.

What would a confidence interval add?

In general, a confidence interval gives a range of values compatible with an estimate under a specified statistical procedure. It communicates uncertainty more directly than a point estimate alone. For a comparison, an interval can help readers assess both the range of plausible effects and whether the estimate is precise enough to distinguish among effects that matter scientifically. No confidence intervals for insulin sensitivity are reported in the registry record, so the uncertainty around any arm difference is unavailable.

Why does non-randomized allocation matter?

In general, non-randomized assignment can leave groups different before an intervention begins. Such differences may relate to the choice or receipt of an intervention and may also be related to the outcome. As a result, a simple comparison of group means may combine intervention-associated changes with pre-existing differences or other influences. The registry record identifies non-randomized allocation and says the arms are not a like-for-like control comparison. It does not report the allocation process, baseline balance, or an adjustment method, so the extent to which these issues were addressed is not reported in the registry record.

What does a baseline-to-follow-up comparison require?

In general, measurements at baseline and follow-up can be used to describe change over time. A sound comparison needs to clarify how repeated measurements were handled, which participants contributed data at each time point, and how missing observations were treated. A within-arm change is not the same as a difference between arms. Comparing changes also does not remove bias from non-randomized allocation by itself. The registry record names the time points but does not report these analytic details or the resulting estimates.

What do the planning assumptions mean?

The record describes a sample-size planning illustration based on the observed effect, with a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is described as a stated fraction of the observed effect. These are planning assumptions, not evidence about the study’s actual statistical power or a replacement for its own design justification. The numerical planning results and the observed effect used for that illustration are not reported in the registry record. Such sample-size illustrations are planning estimates, never observed power.

In general, over-dispersion refers to variability greater than a statistical model expects. It can affect standard errors and the reliability of inferential results when the model does not represent the data’s variability. That concept is not a reported feature of this continuous endpoint analysis; the registry record does not report a model or analysis from which over-dispersion could be evaluated.

More broadly, an analysis plan would ordinarily clarify the primary estimand, the analysis population, how repeated measurements and missing data are handled, and whether adjustments are made for baseline values or multiple outcomes. These details are not reported in the registry record. With 0 analyses listed and no endpoint estimates, there is no basis to reconstruct a statistical test or to calculate a separate result. The endpoint definition and planning assumptions describe measurement and planning, not findings.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists insulin sensitivity as the primary outcome and Body Weight and Fitness as secondary outcomes. It reports 3 outcome measures and 0 analyses. These details identify a set of measurements, but they do not describe how the outcomes were analyzed or how inferences across outcomes were handled.

When a study evaluates multiple outcomes or multiple comparisons, general statistical practice considers the possibility that some apparently notable findings may occur by chance. A prespecified hierarchy, adjustment method, or other strategy can clarify how to interpret several tests. The registry record does not report a multiplicity strategy, so it is not possible to determine whether any adjustment was planned or applied.

The record does not report the number of participants contributing endpoint data at each time point, the amount or pattern of missing data, or a method for handling missing observations. The analysis population is also not reported in the registry record. Therefore, it is unclear from the record which participants were included in any outcome estimate, and there is no basis for assessing how missing observations might affect results.

Similarly, baseline adjustment, repeated-measures methods, covariates, and sensitivity analyses are not reported in the registry record. These are not assumed to have been absent from the full study; their status in the registry record is simply not reported. With no listed analyses or numerical outcome results, neither multiplicity nor missing-data methods can be evaluated, and no inferential conclusion should be attributed to the registry record.

11. 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 0% reporting threshold.

Most frequent non-serious adverse event terms

The registry record reports serious adverse event counts of 0 for Arm 1 and 0 for Arm 2. The corresponding numbers at risk are 59 in Arm 1 and 60 in Arm 2. These are the recorded counts for the reported safety categories; they do not establish that no adverse events of any kind occurred. The record reports one other adverse event for Arm 2 and zero for Arm 1, with the same respective numbers at risk.

In general, an observed count of zero does not establish that an event is impossible or that the underlying risk is zero. Safety counts also need to be interpreted in light of the observation period, how events were collected and classified, and how many participants contributed safety data. The registry record does not report further details about event collection or follow-up for these figures.

Deaths are not reported in the registry record. No serious adverse event terms are listed. The available counts do not support a broad comparison of safety between the interventions, particularly given the non-randomized allocation and the absence of additional safety analyses. They should be read as registry-reported counts, not as a complete characterization of risk.

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

The study record describes a comparison of 6 months of aerobic exercise training with 6 months of weight loss, using insulin sensitivity as the primary endpoint. It establishes the research question and the measurement approach, including a hyperinsulinemic-euglycemic clamp and measurements at baseline and 6 month. It does not report the numerical primary endpoint results.

Accordingly, the registry record does not show whether the mean insulin-sensitivity measure increased, decreased, or remained similar in either arm. It does not show whether changes differed between arms, how precise any estimate was, or whether a statistical test supported a difference. The listed secondary outcomes, Body Weight and Fitness, also have no numerical results reported in the registry record.

The study’s non-randomized allocation is important to the interpretation. Even if outcome estimates were available, a difference between arms would not automatically represent an effect caused by the intervention. Group differences and other factors could contribute, and the registry record does not report an adjustment strategy or enough analytic detail to evaluate that possibility.

The background rationale refers to chronic inflammation, insulin resistance, and lifestyle interventions. These are motivations for the research, not trial findings. The record does not report a measured inflammatory outcome result or evidence linking a change in inflammation to a change in insulin sensitivity.

In general, a result is most informative when its estimate, uncertainty, analysis population, and analytic approach are all available. Those elements are not reported for the primary endpoint in the registry record. The defensible conclusion is limited to what the record documents: a completed non-randomized study with defined interventions and an insulin-sensitivity endpoint, but without posted outcome estimates that would permit evaluation of the interventions’ comparative results.

13. Why This Trial Matters Statistically

This study is statistically notable for the distinction between having a clearly described endpoint and having interpretable outcome evidence. The registry record defines insulin sensitivity, specifies a measurement method, identifies the measurement times, and states how the continuous endpoint is summarized. These details help readers understand what the study intended to measure. They do not substitute for results, and the record does not report the endpoint estimates or their uncertainty.

The non-randomized design is another central statistical feature. It makes the comparison different from a randomized trial, because allocation does not ensure that the intervention arms are comparable at baseline. This matters when interpreting any apparent difference: statistical adjustment may address some measured differences, but the registry record does not report whether such adjustment was used, and unmeasured differences can remain a concern in non-randomized comparisons.

The study also illustrates why the registry’s analysis information matters. It lists 3 outcome measures and 0 analyses. Without an analysis description, readers cannot determine how baseline and follow-up measurements were compared, what analysis population was used, or how missing observations and multiple outcomes were handled. Those omissions limit the ability to assess both the strength and the uncertainty of any conclusions.

Finally, the recorded safety counts should be kept distinct from efficacy interpretation. They report event counts, but do not provide a comprehensive account of safety or a randomized comparison. Reading design, endpoint definitions, analyses, and safety information separately helps avoid treating a rationale or an outcome description as if it were evidence of an effect.

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 gives limited information for evaluating the study’s comparative results. It identifies non-randomized allocation and explicitly cautions that the arms are not a like-for-like control comparison. It does not report how participants entered each arm, what baseline characteristics were measured, or whether the groups were balanced. This leaves important potential sources of confounding unaddressed in the posted information.

The primary endpoint is described, but its numerical results are not reported in the registry record. The record also does not report estimates of change, uncertainty intervals, p-values, or the results for the listed secondary outcomes. With 0 analyses listed, the statistical approach used to compare measurements cannot be assessed.

Information about missing observations and analysis populations is not reported in the registry record. The record does not state how many participants contributed measurements at baseline or 6 month, whether observations were incomplete, or how any missing values were handled. It also does not give details about adherence to the interventions or other factors that might affect interpretation.

The study’s background discusses inflammation, but the listed primary endpoint is insulin sensitivity, and an inflammation outcome result is not reported in the registry record. The safety information is also limited: serious adverse event counts and other adverse event counts are listed, but deaths are not reported, and the record does not provide further event-collection details.

These limitations concern what can be learned from the registry record. They do not establish that relevant methods or data were absent from the full study; those details are not reported in the registry record.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Exercise and Inflammation in the Elderly — NCT00753363.
• Registry record last updated 2018-06-26; status last verified 2018-03.
• Results first posted 2014-12-22.
• 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, parallel, non-randomized study sponsored by the VA Office of Research and Development. It reports enrollment of 119, a start date of 2007-04-01, and a primary completion date of 2012-11-29.

The two arms involved 6 months of aerobic exercise training and 6 months of weight loss. The primary endpoint was insulin sensitivity, measured at baseline and 6 month using glucose utilization during a hyperinsulinemic-euglycemic clamp. Body Weight and Fitness are listed as secondary outcomes.

Numerical outcome results and analyses are not reported in the registry record. The record lists 3 outcome measures and 0 analyses. It reports zero serious adverse events in each arm, one other adverse event for Arm 2, and zero for Arm 1. Deaths are not reported in the registry record. Because allocation was non-randomized, the arms are not a like-for-like control comparison.