Multilevel Intervention for Physical Activity in Retirement Communities: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Multilevel Intervention for Physical Activity in Retirement Communities (NCT01155011), a randomized na study. The primary outcome is Daily Minutes of Physical Activity.

2. Clinical Question

The registry describes a study of sedentary adults aged 65 and older living in continuing care retirement communities. Its stated purpose was to assess whether a 6-month multilevel physical-activity intervention could increase physical-activity levels. The primary outcome listed in the results, however, is daily minutes of physical activity at baseline, before the intervention period could be evaluated as a change over time.

The registry records a randomized, parallel-group study, with the community site as the unit of randomization. The posted baseline means describe the groups at the outset; they should not be read as a post-intervention comparison or as an estimate of the intervention's effect. A post-intervention primary outcome or a change-from-baseline result is not reported in the registry record.

3. Trial Design

The study is identified as MIPARC, a randomized, parallel-group trial with triple masking recorded in the registry. The registry lists 307 participants and describes a group-randomized design in which the site, rather than an individual resident, was the unit of randomization. The study summary reports that residents in 11 continuing care retirement communities received either MIPARC or an active health-education program for 6 months.

The intervention was multilevel. Its listed components include group educational sessions, phone counseling, peer mentoring, policy change, support, and tailored environmental resources. The intervention description also mentions pedometer self-monitoring and educational materials, although the registry text is truncated. The control was not described as no treatment: it received health-education lectures and mailed materials, with lectures scheduled to match the MIPARC schedule.

The primary outcome listed in the results is daily minutes of physical activity measured by 7 day accelerometry with a 760 cpm cutpoint. Its timeframe is Baseline. The registry reports a mean and standard deviation for each arm at that time, along with an unadjusted mean difference and confidence interval. Those baseline figures characterize the randomized groups at study entry; they do not quantify an intervention-period outcome.

The registry lists the study as completed, with a start date of 2011-01 and a primary completion date of 2014-07. The phase is recorded as NA. Details of the original allocation procedure beyond site-level randomization, the masking procedures, the original sample-size justification, and the planned statistical analysis are not reported in the registry record.

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.

Baseline to 6 Months

Reasons for not completing this period:

6 to 12 Months

Reasons for not completing this period:

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 Daily Minutes of Physical Activity, measured over Baseline.

The primary outcome is daily minutes of physical activity, a continuous measure summarized as a mean and standard deviation. The registry says it was measured by 7 day accelerometry using a 760 cpm cutpoint. Accelerometry records movement objectively rather than relying only on participants' recollection, while the specified cutpoint determines which recorded activity counts toward the measure. The registry does not provide further detail about accelerometer wear rules or data processing.

Minutes per day provide an interpretable measure of the amount of activity captured by the device under the stated cutpoint. A mean describes the group's average activity, while the standard deviation describes how spread out individual measurements are around that average. Together, these summaries convey both the central level and variability of activity in each arm.

The listed timeframe is Baseline. That timing is essential: the posted means of 50.03 minutes in the MIPARC arm and 39.19 minutes in the health-education control arm describe activity at study entry. Although the study's stated aim concerns an increase over 6 months, baseline values alone cannot show whether activity subsequently changed, whether any change differed between groups, or whether an observed difference arose after intervention exposure. A follow-up value or a change analysis is not reported in the registry record.

7. Results Posted in the Registry

• MIPARC Intervention: mean 50.03 versus 39.19 on control; difference 10.84 (95% CI 4.43 to 17.25), standardized effect d = 0.38; z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous: daily physical-activity minutes. For a continuous measure, a group mean summarizes the average observed value, and a standard deviation describes the variability of individual values around that average. The registry lists 50.03 minutes with a standard deviation of 29.02 in the MIPARC arm, and 39.19 minutes with a standard deviation of 28.3 in the control arm. These summaries relate to baseline, not to the end of the intervention.

What does the mean difference describe?

The reported mean difference is 10.84 minutes, calculated as the MIPARC baseline mean relative to the control baseline mean. A difference is expressed in the same units as the outcome, so its interpretation remains in daily minutes. The unadjusted 95% confidence interval is 4.425 to 17.25 minutes. The interval quantifies uncertainty in this summary comparison under the calculation's assumptions; it is not an interval for an intervention effect at follow-up.

The registry also reports p < 0.001 for the unadjusted comparison. A p-value measures how inconsistent the observed statistic would be with a specified null model, assuming the model and analysis assumptions hold. It is not the probability that the null hypothesis is true, and it does not measure clinical importance. Most importantly, the null comparison represented by this posted statistic is a baseline mean comparison, not a test of the planned 6-month increase.

Why does the allocation unit matter?

The study summary says sites were randomized. This is cluster randomization: residents are grouped within communities, and a community is assigned as a whole. People in the same community may have more similar activity because they share facilities, schedules, and local influences. That resemblance is called within-cluster correlation. If an analysis ignores it, it can treat correlated observations as more independent information than they are, making standard errors too small and confidence intervals too narrow.

Cluster-aware methods can account for this dependence, for example by modeling community-level variation or adjusting uncertainty for clustering. The registry does not report the method used for the sponsor's analysis, the number of sites assigned to each arm, a cluster-level correlation estimate, or whether the posted unadjusted comparison accounts for clustering. The posted summary should not be assumed to be cluster-adjusted. The original analysis details are not reported in the registry record.

What does the standard deviation add?

The standard deviations of 29.02 and 28.3 minutes indicate substantial variation in individual baseline activity measurements within the arms. A mean difference can be estimated even when individual measurements vary widely, but that variation affects precision. The standard deviation is not the same as the standard error: the former describes variation among observations, while the latter describes uncertainty in an estimated mean or contrast. The registry lists standard errors of 2.362 and 2.266 for the two arm means.

Accelerometer outcomes also depend on measurement definitions. The registry specifies a 7 day measurement period and a 760 cpm cutpoint, but wear-time requirements, handling of incomplete recordings, and other processing details are not reported in the registry record. Those details can affect which observations are usable and what activity is counted.

How should the inference be bounded?

The numerical summaries are useful for describing baseline activity and the uncertainty of the unadjusted contrast. They cannot answer whether activity increased during treatment. A follow-up contrast could be assessed as an end-of-period difference, a change from baseline, or a group-by-time comparison, but the registry does not report which analysis was planned or any such result. These approaches are not interchangeable in every setting; their interpretation depends on the design, measurement times, and analysis model.

General statistical principle: inference should match both the randomization unit and the question being asked. For a site-randomized intervention, uncertainty should reflect clustering, and an intervention claim should rely on measurements that capture outcomes after exposure. Because the posted primary result is baseline and the analysis plan is not reported, the unadjusted figures support a descriptive baseline comparison rather than a definitive treatment-effect estimate.

10. Confidence Intervals and Precision

The unadjusted mean difference in baseline daily physical-activity minutes is 10.84. Its 95% confidence interval extends from 4.425 to 17.25 minutes. The interval gives a range of values that are compatible with the observed data under the calculation and its assumptions. It communicates uncertainty more directly than a p-value alone, while retaining the same limitation: it describes the baseline contrast, not a 6-month intervention effect.

The interval does not include zero, and the reported unadjusted p-value is p < 0.001. This indicates that the baseline contrast is statistically distinguishable from zero under the unadjusted comparison model. Statistical distinguishability does not determine whether a difference matters practically, and it cannot establish that the intervention produced the difference. The measurement occurred at baseline.

Precision also depends on the design and analysis. The communities were randomized as groups, so within-community correlation may affect the uncertainty of a participant-level contrast. The registry does not report whether the interval accounts for clustering. Without that information, the displayed interval should not be interpreted as confirmed cluster-adjusted precision. The listed standard deviations, 29.02 and 28.3, also show individual variability, but they do not by themselves establish the precision of a properly design-adjusted treatment estimate.

A narrow interval can still answer the wrong timing question. Here, the reported interval concerns the difference between the arm means at baseline. A confidence interval for an intervention effect would need a follow-up or change-based estimate and an analysis appropriate to the site-randomized design. Such an estimate is not reported in the registry record.

11. 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 planning figures are participants per arm, not total event counts. They are illustrations based on the observed mean difference, not the trial's original sample-size calculation. The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The effect to detect is set to a stated fraction of the observed effect. The registry does not report the original design assumptions or the trial's sample-size justification.

For planning around 100% of the observed effect, the illustration gives 110 participants per arm for 80% power and 147 participants per arm for 90% power. For planning around 75% of the observed effect, it gives 196 participants per arm for 80% power and 262 participants per arm for 90% power. For planning around 50% of the observed effect, it gives 439 participants per arm for 80% power and 588 participants per arm for 90% power.

In plain terms, power is the probability that a study would meet its statistical detection criterion if the specified effect were the true effect and the planning assumptions held. The 80% and 90% columns represent two chosen power targets; the higher target calls for more participants under otherwise identical assumptions. Smaller effects are harder to distinguish from random variation, so planning to detect a smaller fraction of the observed effect requires more participants per arm.

These are planning illustrations, not observed power. They do not show what power this completed trial actually achieved, and the observed baseline difference should not be treated as a proven intervention effect. The illustrations also state equal allocation and a normal approximation; they do not report a cluster-randomized adjustment. The trial's site-based allocation could affect a real design calculation, but its original handling is not reported in the registry record.

12. 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 = 10.840; pooled standard deviation σ = 28.656.
2. n = 2 × (1.96 + 0.842)² × 28.656² / 10.840² = 109.7, rounded up to 110 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 <- 10.84                      # observed difference in means
sigma <- 28.6563591203207                     # 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(1155011)
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 = 110 (80% power), 147 (90% power)
 75% of observed effect: n per arm = 196 (80% power), 262 (90% power)
 50% of observed effect: n per arm = 439 (80% power), 588 (90% power)
power.t.test: 110.7 
Simulated power at n = 110 per arm: 0.801 (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.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists three outcome measures, but no secondary outcome results are posted. It also records zero analyses. The available results therefore do not describe a complete set of planned outcome comparisons, and the registry record does not report which outcomes or time points were included in the original analysis plan.

Multiplicity arises when several outcomes, time points, or statistical comparisons are examined. With more opportunities to find a small p-value, the chance of at least one apparently notable result can increase unless the analysis plan addresses this issue. The registry does not report whether a multiplicity strategy was specified, such as selecting a single primary comparison or adjusting inference across multiple comparisons. Because the posted primary outcome is baseline activity, this concern does not change the central interpretation: the posted result is not a follow-up intervention-effect estimate.

Missing-data handling is also not reported in the registry record. The record does not state how many measurements were unavailable at baseline or follow-up, how incomplete accelerometer recordings were handled, whether participants were excluded from analyses, or whether missing values were imputed. Arm-level outcome counts are listed for the baseline measure, but there is no follow-up result from which to assess retention or missingness during the intervention period.

The analysis population is not described. The registry does not say whether analyses followed an intention-to-treat principle, included only participants with usable measurements, or used another population definition. It also does not report whether any participants changed exposure or whether such changes were handled in the analysis. These details matter because exclusions or missing measurements can affect comparability created by randomization, especially when the intervention is assigned at the community level.

Finally, the record does not report a statistical analysis plan, model specification, or handling of community clustering. These absences limit the ability to assess whether the posted inference aligns with the trial design and planned outcomes. They should be treated as unreported methods, not as evidence that a particular method was or was not used.

14. 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 serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: fall (13.2%), overnight hospitalization (6.6%), serious physical injury (3.3%), death (3.2%), persistent or significant physical or cognitive disability or incapacity (2.0%).

The registry reports serious adverse events affecting 38 participants in the MIPARC arm and 30 in the health-education control arm. Other adverse events affected 23 and 12 participants, respectively. The figures at risk are listed as 151 for MIPARC and 156 for control. The registry also lists deaths of 1 and 5, respectively. These are descriptive counts; they do not establish that the intervention caused an event or prevented one.

Among the listed serious-event terms, the registry gives Fall at 13.2%, Overnight hospitalization at 6.6%, Serious physical injury at 3.3%, Death at 3.2%, and Persistent or significant physical or cognitive disability or incapacity at 2.0%. The denominator and counting rules for these term percentages are not reported in the registry record. The listed Death term percentage should not be treated as interchangeable with the separate arm-level death counts.

Adverse-event counts need context, including how events were defined, collected, attributed, and whether repeated events in one participant were counted once or more than once. Those details are not reported in the registry record. Without them and without an inferential safety analysis, the counts should be read as registry summaries rather than as comparative evidence of safety.

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

The posted outcome shows that the MIPARC group had higher mean daily physical activity at baseline than the health-education control group: 50.03 versus 39.19 minutes. The unadjusted mean difference is 10.84 minutes, with a 95% confidence interval from 4.425 to 17.25 and p < 0.001. This is a description of the baseline groups and the uncertainty around their unadjusted difference.

It does not show that MIPARC increased physical activity over 6 months. The primary outcome timeframe is Baseline, so the posted values precede the intervention period described in the study summary. No post-intervention activity result, change from baseline, or other between-group estimate of change is reported in the registry record. A baseline difference cannot substitute for such evidence, even when its p-value is small.

The result also does not show whether the difference is important to residents' functioning or health, whether it persisted, or whether it was attributable to particular intervention components. The registry does not report a cluster-adjusted analysis, and the site-level randomization makes the handling of community clustering relevant to uncertainty. Those limitations constrain interpretation of the confidence interval and p-value.

The trial's stated question remains scientifically distinct from the comparison that can be read from the posted results. Answering whether the intervention changed activity requires appropriately timed outcome data and analysis that reflects the group-randomized design. The baseline summaries can characterize the enrolled arms, but they cannot establish the direction or size of a treatment effect.

16. Why This Trial Matters Statistically

This trial is statistically instructive because its allocation unit and its outcome summaries operate at different levels. The study randomized communities, while the posted primary outcome is summarized across residents. In cluster-randomized studies, outcomes among people in the same site may be correlated. Analysis and uncertainty estimates need to reflect the clustering rather than automatically treating every resident as independent.

It also illustrates why outcome timing is central to interpretation. The study's stated purpose concerns physical activity over a 6-month intervention, but the posted primary measure is explicitly at baseline. A numerical contrast and a small p-value can look persuasive while addressing a different question from the intended treatment-effect question. Statistical significance cannot repair a mismatch between measurement time and causal claim.

The active control is another relevant design feature. Both arms received an organized program, while the control addressed general health and healthy aging. Any valid estimate of the program's effect would therefore concern the contrast between MIPARC and this health-education condition, not the contrast with receiving no intervention. The posted baseline comparison does not estimate either version of that effect.

Finally, the record illustrates the value of reporting analysis details alongside results. Follow-up outcomes, handling of clustering and missing data, analysis populations, and multiplicity procedures would help readers judge whether inference matches the design and question. These details are not reported in the registry record, which limits what can be concluded from the posted statistics.

17. 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 most consequential trial-specific limitation of the posted results is timing: the primary outcome is recorded at Baseline, although the study summary describes a 6-month intervention. The registry does not report a follow-up activity outcome or change estimate, so the intervention question cannot be answered from the posted primary result.

The study randomized sites, and the summary reports 11 continuing care retirement communities. The registry does not report how many communities were assigned to each arm, the cluster-level analysis method, or whether the posted comparison adjusts for correlation within communities. A participant-level contrast without appropriate cluster handling may give misleading uncertainty.

Measurement details are incomplete. The record specifies 7 day accelerometry and a 760 cpm cutpoint, but does not report accelerometer wear criteria, valid-day requirements, treatment of incomplete recordings, or processing decisions. These details can affect the observed activity measure. The intervention description is also truncated, limiting assessment of exactly which components were delivered and how consistently.

Other limitations include the absence of posted secondary results, missing-data methods, analysis-population definitions, multiplicity procedures, and a detailed statistical plan. The original sample-size assumptions and justification are not reported in the registry record. The available adverse-event summaries also lack reported definitions and denominator details for the listed serious-event terms. These omissions restrict evaluation of both efficacy and safety; they should not be interpreted as evidence about what methods the investigators used.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Multilevel Intervention for Physical Activity in Retirement Communities — NCT01155011.
• Registry record last updated 2017-08-31; status last verified 2017-08.
• Results first posted 2017-07-31.
• 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.

21. Record Summary

The registry describes MIPARC, a completed randomized parallel-group study of a multilevel physical-activity intervention versus an active health-education control in sedentary residents aged 65 and older living in continuing care retirement communities. The summary identifies site as the unit of randomization and describes an intervention period of 6 months.

The primary outcome is daily minutes of physical activity measured by 7 day accelerometry with a 760 cpm cutpoint. The posted timeframe is Baseline. The reported mean difference of 10.84 minutes, 95% confidence interval of 4.425 to 17.25, and p < 0.001 therefore describe an unadjusted baseline contrast, not a change attributed to MIPARC.

Follow-up primary results, secondary results, and important analysis details are not reported in the registry record. The posted statistics do not establish the intervention's effect.