Reducing Traffic Pollution Exposure Improves Blood Pressure: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Reducing Traffic Pollution Exposure Improves Blood Pressure (NCT04029129), a randomized na study. The primary outcome is Systolic Blood Pressure (SBP).

2. Clinical Question

3. Trial Design

This was a randomized, single-masked, crossover trial of room-level air filtration and building-envelope modifications. Its registry title is “Reducing Traffic Pollution Exposure Improves Blood Pressure,” and its official title is “Near Highway Pollution: From Research to Action.” The record lists UConn Health as the sponsor, a start date of 2018-03-01, a primary completion date of 2018-06-15, and a completed status. The phase is listed as not applicable.

The crossover structure means that exposure conditions were compared within a trial organized around different conditions rather than as a simple parallel-group design. The registry record does not describe the order in which participants experienced the conditions, how that order was assigned, whether a washout period was used, or how period and carryover effects were handled. It also does not identify who was masked. These details matter when interpreting comparisons because observations from the same participant across conditions may be related.

The high-exposure room allowed ambient air to enter freely. Limited filtration was used for medium exposure, while low exposure involved closed and sealed doors and windows together with full filtration. Each condition was described in relation to reducing pollution in the room; pollution concentrations are not reported in the registry record.

The registry lists enrollment of 77. The posted arm summaries each report 77 observations, but the record does not explain the relationship between these arm counts and the crossover structure. It is not possible from the registry alone to determine whether every participant contributed measurements under every condition or how missing observations were handled. The planned sample size, original design assumptions, randomization procedure, allocation concealment, and analysis model 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.

Before assignment: Using a 3-period crossover design, 77 participants were randomized to attend. Three 2-hour-long exposure sessions separated by 1-week washout periods. None was excluded.

First Intervention (2 Hours)

Wash Out (1 Week)

Third Intervention (2 Hours)

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: A linear mixed model will be used to analyze the 3-exposure, 3-period crossover design. The primary outcome measure is the change in systolic blood pressure from the beginning of each exposure session. Before calculating change scores in each exposure session, each participant's systolic blood pressure measurements will be smoothed using a 3-time-period, centered moving average.

6. Primary Endpoint

The registry lists the primary outcome as Systolic Blood Pressure (SBP), measured over Over 2 hour exposure period, we measured SBP after 20 minutes with the objective to assess change in SBP in a time series of measurements over that time period..

The primary endpoint is systolic blood pressure (SBP), expressed in mmHg and measured with ambulatory blood pressure monitors. SBP is the pressure recorded during the heart’s contraction. Measuring it during controlled exposure conditions offers a way to describe whether the posted blood-pressure summaries vary when room filtration and building sealing are changed.

The registry describes an exposure session lasting two hours and says SBP was measured after 20 minutes, with the objective of assessing change in a time series of measurements over that period. This wording indicates an interest in the pattern of SBP over time, not only a single isolated reading. However, the posted results provide arm-level means and standard errors rather than a time-series analysis. The registry record does not clarify whether those means represent a particular time point, an average across measurements, or another summary of the repeated readings.

Using the same outcome measure across exposure conditions makes it possible to describe contrasts in the observed SBP summaries. In a crossover study, the analysis may also need to account for the relationship between measurements from the same person under different conditions. The registry record does not report an analysis model that explains how this was handled. The posted means are therefore descriptive summaries; they should not be mistaken for a complete account of the time-varying primary endpoint.

7. Results Posted in the Registry

• Medium Exposure: mean 120.83 versus 122.30 on control; difference -1.47 (95% CI -3.55 to 0.61), standardized effect d = -0.22; z-test p = 0.167.
• Low Exposure: mean 119.90 versus 122.30 on control; difference -2.40 (95% CI -4.62 to -0.18), standardized effect d = -0.34; z-test p = 0.034.

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.

Diastolic Blood Pressure (DBP)

• 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:
  • Medium Exposure: mean 70.40 versus 69.90 on control; difference 0.50 (95% CI -0.89 to 1.89), standardized effect d = 0.11; z-test p = 0.480.
  • Low Exposure: mean 71.00 versus 69.90 on control; difference 1.10 (95% CI -0.29 to 2.49), standardized effect d = 0.25; z-test p = 0.120.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Medium Exposure and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Low Exposure and the control. The interval gives the range of effects that remain compatible with the data.

The registry posts diastolic blood pressure (DBP) as a secondary outcome. Across the exposure conditions, its pattern does not mirror the primary SBP pattern: the DBP summaries are higher in the medium- and low-exposure conditions than in high exposure, whereas the SBP summaries are lower in those conditions. The posted unadjusted DBP comparisons have intervals that cross zero. This broad contrast in direction is a reason to avoid treating the primary and secondary outcomes as interchangeable measures of one uniform response.

These observations are descriptive, and the secondary results do not establish that the outcomes respond differently. SBP and DBP measure distinct components of blood pressure, and the registry does not report a formal test comparing their exposure effects. Nor does it report a joint model or a prespecified approach for interpreting the outcomes together. Any apparent divergence should therefore be treated as a feature of the posted estimates, not as proof of a biological explanation.

Secondary outcomes generally carry less confirmatory weight than a prespecified primary endpoint, particularly when the analysis plan and multiplicity strategy are not available. Testing multiple outcomes or exposure contrasts increases the opportunity for chance findings if each is considered in isolation. The registry record does not report whether the DBP analysis was adjusted for multiple comparisons, whether it was prespecified, or how missing measurements were handled.

The secondary outcome may help readers understand the range of measurements collected during the exposure sessions, but it cannot independently resolve the uncertainty in the SBP comparison. Its role is best viewed as complementary and exploratory in the absence of a reported analysis plan. The record’s posted summary is not a substitute for an analysis that accounts for the crossover and time-series structure.

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

• Diastolic Blood Pressure (DBP)

9. Statistical Interpretation

10. Statistical Methods Explained

The registry describes a randomized crossover trial, but it does not report the statistical methods used by the sponsor. There are no reported analyses in the record. The numerical contrasts associated with the posted arm summaries are unadjusted recalculations, rather than a documented sponsor model. That distinction is important: an analysis based only on summary means and standard errors cannot represent every feature of a crossover study.

What does a mean difference express?

A mean difference compares the average outcome in one condition with the average in a reference condition. Here, high exposure is the reference for the posted SBP comparisons. A negative difference means that the mean in the compared condition is lower than the high-exposure mean; a positive difference would mean it is higher. This describes direction and size in the outcome’s units, which are mmHg for SBP. It does not by itself say whether the estimate is precise, clinically important, or free from bias.

The SBP mean difference for medium exposure is -1.47 mmHg, and the difference for low exposure is -2.4 mmHg. These are recalculated unadjusted comparisons from the posted summary statistics. Since the registry record does not report a sponsor analysis, readers cannot compare these simplified results with the trial’s own model-based estimates or determine whether the sponsor used a different endpoint summary.

What do standard errors describe?

A standard error describes the uncertainty in an estimated mean under the statistical assumptions used to calculate it. It is not the spread of individual blood pressure readings. The posted arm summaries report standard errors for the means. Standard errors can help convey how precisely the arm means are estimated, but they do not disclose the full distribution of participant measurements, the within-person correlations in a crossover trial, or the variability in change over time.

For repeated measures, the association between measurements from the same person can affect the uncertainty of a contrast. General statistical reasoning is that paired or repeated observations may yield different uncertainty from an analysis that treats conditions as independent. The posted recalculations do not report using a within-person covariance estimate. The registry record does not provide enough analysis detail to establish how the original trial analysis accounted for repeated observations.

How should the intervals be understood?

A confidence interval gives a range of values compatible with the observed estimate under the method’s assumptions. It communicates uncertainty about the contrast, not the range of outcomes for individual participants. The interval for the medium-versus-high SBP comparison crosses zero; the low-versus-high interval lies below zero. The confidence level for the intervals is not reported in the registry record, so it should not be inferred from the posted estimates alone.

Intervals are often more informative than a threshold-based reading of a p-value because they display both the direction and a range of plausible effect sizes. Their interpretation remains conditional on the calculation method and the data structure. In this case, the simple recalculation does not establish whether the interval reflects a crossover analysis, and the registry record does not state the sponsor’s method.

What does an unadjusted p-value mean?

An unadjusted p-value measures how incompatible the observed data are with a specified null hypothesis under the assumptions of the calculation. It is not the probability that the null hypothesis is true, and it does not measure the magnitude or practical importance of a result. The posted p-values for the SBP comparisons are unadjusted. They do not account for the fact that the record includes multiple exposure comparisons and a secondary outcome.

Unadjusted results can be useful descriptive evidence, but their interpretation is more cautious when several questions are considered. A multiplicity adjustment or a prespecified hierarchy can help define how to interpret a collection of tests; neither is reported in the registry record. It is also not reported whether the displayed comparisons were the only comparisons considered by the trial investigators.

Why does the crossover structure matter?

In a crossover study, each participant may experience more than one condition, so outcome measurements can be related within a person. Analyses commonly need to reflect the repeated structure and may need to consider condition order, period, and possible carryover, depending on the protocol. These are general design considerations, not claims that a particular problem occurred in this trial. The registry record does not report the condition sequence, washout details, or an analysis of period or carryover effects.

The primary endpoint description refers to a time series of SBP measurements over the exposure period, while the posted arm results are means. The record does not explain how the time series was reduced or modeled. As a result, the displayed contrasts cannot answer whether an effect changed over the session or whether a particular point in time drove the summary. Those questions require the underlying measurements and a clearly described analysis.

11. Confidence Intervals and Precision

Precision concerns how narrowly the data locate an estimated contrast. The posted SBP interval for medium exposure relative to high exposure crosses zero. It therefore includes both a negative contrast and a positive contrast, rather than isolating one direction. The interval for low exposure relative to high exposure is entirely below zero. The registry record does not state the confidence level used for these intervals, so their numerical bounds should be interpreted with that limitation in mind.

The estimates are not equally precise simply because they come from the same trial. Precision depends on the variability of measurements and the analysis method, including how repeated observations are handled. The arm-level standard errors are 0.8 for high exposure, 0.7 for medium exposure, and 0.8 for low exposure. These summarize uncertainty around each posted mean, but they do not show the individual observations or the within-person covariance relevant to a crossover comparison.

The width of an interval is not a direct measure of clinical importance. A narrow interval can still be centered on a small difference, and a wider interval may include a range of potentially meaningful differences. The registry record does not define a threshold for clinical importance, nor does it provide a precision target for the primary outcome. The posted estimates should therefore be read as statistical descriptions, not as a determination of clinical relevance.

The interval and p-value for the low-exposure contrast offer evidence that the recalculated unadjusted estimate is not centered at zero under that calculation. They do not resolve uncertainty about the estimate’s exact magnitude, the effects of the crossover structure, the multiple comparisons, or the sponsor’s intended analysis. For medium exposure, the interval’s inclusion of zero reinforces that the summary does not clearly distinguish the contrast from no difference under the recalculation. Neither result alone settles the broader scientific question.

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 planning figures are participants per arm, not total events. They are illustrations computed from the observed effect, rather than the original trial’s design calculations; the original sample-size assumptions are not reported in the registry record. The stated planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each scenario asks what sample size would be needed to detect the stated fraction of the observed effect. These estimates do not incorporate a reported crossover analysis model.

The 80% power column means the estimated chance of detecting the specified effect under the planning assumptions if that effect were the true effect. The 90% power column uses a higher target chance of detection under those same assumptions. Power is a property of a planned study under specified assumptions; it is not the probability that a result is true. These figures are not observed power.

Rows that plan for 100% of the observed effect use the full size of the estimated contrast as the target. Planning for 75% or 50% of the observed effect asks about detecting smaller effects. Smaller target effects generally require more participants to distinguish them from random variation. This is why the planning estimates increase as the target fraction decreases.

The figures are scenario-based illustrations and should not be read as evidence that the trial was designed for those targets or that its actual analysis had the stated properties. The registry record does not provide the trial’s own sample-size justification or the assumptions used to plan enrollment. Because the actual study is crossover, these simple two-arm estimates may not reflect the efficiency or requirements of a properly specified within-person design.

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 = -1.470; pooled standard deviation σ = 6.596.
2. n = 2 × (1.96 + 0.842)² × 6.596² / -1.470² = 316.0, rounded up to 317 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 <- -1.47                      # observed difference in means
sigma <- 6.59583201726666                     # 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(4029129)
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 = 317 (80% power), 424 (90% power)
 75% of observed effect: n per arm = 562 (80% power), 753 (90% power)
 50% of observed effect: n per arm = 1,265 (80% power), 1,693 (90% power)
power.t.test: 317 
Simulated power at n = 317 per arm: 0.795 (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. Multiplicity, Missing Data and Analysis Populations

The registry record does not report a statistical analysis plan, a multiplicity strategy, or sponsor-reported analyses. It lists a primary SBP outcome and a secondary DBP outcome, and it includes comparisons of medium and low exposure with high exposure. Because multiple outcome and exposure questions are present, the interpretation of individual unadjusted p-values depends in part on whether the questions were planned, prioritized, and adjusted in advance. Those details are not reported in the registry record.

Multiplicity matters because considering several statistical tests can increase the chance that at least one appears notable under random variation. General statistical approaches include adjusting significance thresholds, controlling an overall error rate, or specifying a hierarchy in which some outcomes are confirmatory and others exploratory. The record does not say whether any such approach was used. The unadjusted p-values should therefore not be treated as if they were adjusted evidence across the full set of outcomes and comparisons.

Missing-data handling is also not reported. The registry lists an enrollment of 77 and arm-level summaries with 77 observations in each condition, but it does not explain whether these counts reflect complete observations under every crossover condition, how unavailable readings were addressed, or whether any participants were excluded from particular analyses. The adverse-event at-risk counts differ from the outcome summary counts, but the record does not explain why. No assumptions about completeness or exclusion should be inferred from these figures.

The analysis population is not described in the registry record. It does not state whether results were analyzed according to randomized condition, whether all randomized participants were included, or whether another population definition was used. The randomization method, allocation concealment, protocol deviations, and treatment of repeated measurements are also not reported. These omissions limit assessment of bias and reproducibility; they do not establish that any specific problem occurred.

15. 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.

The registry records zero serious adverse events, zero other adverse events, and zero deaths in each of the three exposure conditions. The listed numbers at risk are 30 for high exposure, 28 for medium exposure, and 19 for low exposure. These are the denominators associated with the adverse-event reporting, and they are not the same as the outcome-summary counts.

The figures indicate that no events were recorded in the listed categories among those at risk in each condition. They do not establish that no unrecorded events occurred, or that the conditions have identical safety profiles. The registry record does not explain why the numbers at risk differ by condition, what period of observation was used for adverse events, or how events were collected and classified. The absence of reported events should therefore be read as a feature of the posted safety record, not as a definitive estimate of risk.

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

The posted SBP means are lower under medium and low exposure than under high exposure, and the recalculated low-versus-high contrast has an interval entirely below zero. The medium-versus-high interval crosses zero. These are statistical descriptions of the registry’s summaries; they do not establish that changing building filtration will produce a particular blood-pressure response in other settings or populations.

The randomized crossover design is relevant to the question because participants may experience more than one exposure condition. However, the registry does not describe the condition order, washout, period effects, or sponsor’s analysis. The recalculated contrasts use arm-level summary statistics and do not report adjustment for within-person correlation. Consequently, the posted comparisons cannot fully represent the design unless the calculation method is known to be appropriate for it. The registry record does not provide enough detail to make that determination.

The title makes a directional claim, but the title itself is not evidence. A statistical result needs to be interpreted using the outcome definition, design, analysis, uncertainty, and completeness of reporting. Here, the SBP summaries point in a consistent direction across exposure contrasts, but one interval includes zero and the other does not. The confidence level is not reported, the p-values are unadjusted, and no sponsor analysis is posted. The results therefore warrant a cautious reading rather than a simple yes-or-no conclusion.

The findings also concern short exposure sessions and a room-based intervention. The registry describes a two-hour exposure period, but it does not report measured pollution levels or the detailed time course of the posted outcome means. The data in this record do not establish the effect of longer exposures, effects in other environments, or effects on outcomes beyond the blood-pressure measures listed. Those questions are not answered by the posted summaries.

Overall, the registry record offers a useful descriptive comparison of SBP across three room conditions and posts a secondary DBP outcome. It does not provide the detailed protocol, analysis plan, or sponsor results needed to judge every design and analysis feature. The numerical estimates should be read as conditional summaries, not as medical advice or a guarantee of an individual response.

17. Why This Trial Matters Statistically

This trial raises statistical issues that are common in environmental exposure research. The exposure conditions are created by changing room ventilation and filtration, while the outcomes are measured repeatedly during a short session. The resulting data can have a layered structure: measurements across time, multiple conditions, and potentially repeated observations from the same participant. A clear account of how those layers were analyzed is important for understanding uncertainty.

The crossover design can be informative because participants may serve as their own comparison across conditions, reducing the influence of stable differences between people. That potential advantage depends on the actual protocol and analysis. Condition order, period effects, carryover, and within-person correlation can all matter. The registry record does not report these details, making it impossible to assess how the analysis used the crossover structure.

The comparison also illustrates why effect estimates and uncertainty should be considered together. The direction of the SBP means is consistent across the posted contrasts, but their intervals do not convey the same degree of separation from zero. A p-value alone does not communicate the likely range of an effect or its clinical importance. The absence of a reported confidence level and sponsor analysis adds further limits to interpretation.

Finally, the record highlights the difference between a result summary and a reproducible statistical report. Information on the analysis population, missing readings, multiplicity, time-series modeling, and sample-size justification would help readers understand what was planned and how the posted findings arose. These reporting details are especially important when several outcomes and exposure comparisons are available.

18. 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 leaves several trial-specific uncertainties that affect interpretation. It does not report the randomization procedure, allocation concealment, the sequence of exposure conditions, or the duration and use of any washout. Since the design is crossover, the absence of information on sequence and period handling prevents assessment of whether order or carryover could have influenced the estimates. These are unreported design details, not evidence that a particular bias occurred.

The masking field says single-masked but does not identify who was masked. The room conditions differ in how doors, windows, and filtration were used, but the record does not explain how masking was maintained. It also does not provide measured pollutant concentrations in the room, so the exposure contrast is described by procedures rather than quantitative pollution levels.

The primary endpoint description refers to a time series, yet the posted results give means and standard errors without explaining how readings over the session were summarized. The registry does not report the analysis model, the within-person covariance, or whether period and time effects were included. The comparison estimates are recalculated from summary statistics and should not be mistaken for an analysis that necessarily respects the crossover structure.

Other unreported items include the original sample-size justification, analysis population, missing-data method, multiplicity plan, and sponsor-reported results. The registry lists outcome summaries for 77 observations per exposure condition, while adverse-event at-risk counts vary; the reasons for this are not reported. No assumptions about missing data or participant flow can be made from the available record.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Reducing Traffic Pollution Exposure Improves Blood Pressure — NCT04029129.
• Registry record last updated 2023-09-28; status last verified 2023-09.
• Results first posted 2023-09-28.
• 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.

22. Record Summary

The registry describes a completed randomized, single-masked crossover trial sponsored by UConn Health, with enrollment of 77. It compared high, medium, and low room exposure conditions created through ambient air access, limited filtration, or sealed doors and windows with full filtration. The primary outcome was SBP in mmHg, with DBP listed as a secondary outcome.

The posted SBP means are 122.3 mmHg for high exposure, 120.8 mmHg for medium exposure, and 119.9 mmHg for low exposure. Recalculated unadjusted contrasts show an interval crossing zero for medium versus high exposure and an interval below zero for low versus high exposure. The registry record reports no sponsor analyses and does not provide the analysis plan, crossover details, or multiplicity approach.