Population Based Lighting Study on Older Adults: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Population Based Lighting Study on Older Adults (NCT05676086), a randomized na study. The primary outcome is Change in Pittsburgh Sleep Quality Index (PSQI).

2. Clinical Question

3. Trial Design

The registry describes a randomized, parallel-group study with no masking. The study is marked completed. Enrollment is reported as 191, and the arm-level results list 98 participants in the indoor light intervention group and 93 in the control group. These are the posted arm figures; the registry does not explain the relationship between enrollment and the numbers included in each outcome analysis.

The intervention was a lamp for home use. The comparator received no indoor light supplementation and underwent assessment procedures only. The registry classifies both interventions as behavioral. Details such as the lamp’s technical characteristics, the exposure instructions, the duration of daily use and adherence are not reported in the registry record. The study’s timeframe for the primary outcome is from baseline to the end of the light intervention period, described as 16 weeks.

The study is listed with phase NA, and the registry identifies Azienda Usl di Bologna as sponsor. Its broader project summary describes evidence collection about indoor and outdoor lighting, health, wellbeing and socio-economic factors across pilot cities. That project context does not establish that all of those aims are evaluated by the primary endpoint reported for this trial.

The registry lists a primary PSQI outcome and secondary measures concerning mental and physical health and circadian rest-activity measures. It does not report the detailed randomization procedure, allocation concealment, eligibility criteria, participant flow, analysis population or reasons for any missing outcome data. Those omissions limit how fully the conduct and representativeness of the comparison can be assessed from the registry record alone.

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:

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 Change in Pittsburgh Sleep Quality Index (PSQI), measured over from baseline to the end of the light intervention period - 16 weeks.

The primary endpoint is change in Pittsburgh Sleep Quality Index (PSQI) score from baseline to the end of the light intervention period, described as 16 weeks. PSQI is a measure of sleep quality. The registry describes scores from 0, associated with better sleep, to 21, associated with worse sleep. A score <= 5 is associated with good sleep quality, while a score > 5 is associated with poor sleep quality.

The outcome is useful for this question because the intervention hypothesis concerns sleep quality, and the PSQI provides a structured score for comparing reported sleep status over time. The registry defines a negative difference between baseline and follow-up as improvement: a decrease in score indicates better sleep quality under the stated interpretation. A change score therefore expresses how the score moved over the intervention period, rather than only describing a follow-up score in isolation.

The posted arm-level summaries report a median change of 0.0 for both groups. The corresponding inter-quartile ranges are -1.0 to 2.0 for the indoor light intervention group and -2.0 to 1.0 for the control group. The registry’s reported regression analysis, however, is described as a mean difference in final values. That distinction matters: change from baseline and final value are not the same summary, and a median is not the same measure as a mean. Interpretation should retain those differences rather than treating the estimates as duplicate presentations of one statistic.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

General: understanding the summaries. A median is the central value when observations are ordered. It is less affected by extreme observations than a mean, but it does not describe how far each observation lies from the center. An inter-quartile range describes the span from the lower to the upper quartile, covering the middle portion of a distribution. Here, the arm summaries use the median and inter-quartile range for PSQI change. They are descriptive statistics, not by themselves a test of whether the intervention and control groups differ.

The posted median change is 0.0 in each arm, but the ranges differ. That does not establish that the distributions are identical: the inter-quartile ranges describe only part of the distribution, and they do not show every score or every participant’s trajectory. The registry does not post a distributional plot or a test based on the two median changes within the registry entries described for the primary result. More broadly, a measure of center and a measure of spread answer complementary questions: the center describes a typical value, while the spread indicates variability around it.

General: interpreting the regression estimate. Linear regression models an outcome as a function of predictors and estimates a contrast conditional on the model specification. The registry reports the method as linear regression and describes the parameter as a mean difference in final values. It does not report the full model specification, any covariates, assumptions checks or diagnostic results. It is therefore possible to explain what a regression mean difference generally represents, but not to reconstruct the model or assess those details from the registry record.

The reported mean difference is 0.52. Its 95% confidence interval runs from -0.18 to 1.2. A confidence interval is a range of estimates compatible with the observed data under the model and the procedure used to calculate the interval. In repeated use of that procedure under its assumptions, intervals would have their stated coverage. It is not a statement that there is a 95% probability that this particular interval contains the true value. Because the posted interval includes zero, it does not establish a nonzero difference at the corresponding conventional two-sided level.

The reported p-value is 0.14. A p-value measures how incompatible the observed data are with a specified null model, under the assumptions of the analysis. It is not the probability that the null hypothesis is true, nor does it measure the size or clinical importance of an effect. The registry reports a superiority analysis, but does not report the detailed hypothesis, test statistic, sidedness or other inferential procedures. The p-value should therefore be read as part of the sponsor’s reported analysis, with those methodological details not reported in the registry record.

General: why estimate and uncertainty both matter. An effect estimate gives the direction and scale of a contrast, while an interval communicates its precision. A result can be statistically non-significant while still being compatible with a range of possible effects. A failure to reject a null hypothesis is not evidence that the groups are equivalent. Equivalence or non-inferiority requires a prespecified margin and a suitable design and analysis; such a margin is not reported in the registry record.

The current summaries require care because the arm figures describe median change, whereas the regression analysis describes a mean difference in final values. A change score incorporates baseline and follow-up into a within-person difference; a final-value comparison concerns the outcome at follow-up. These can yield different estimates, particularly when baseline values vary or when modeling choices differ. The registry does not report the final-value means by arm, baseline values, a baseline-adjusted model description, or a regression estimate for change scores. No direct numerical reconciliation can be made from the reported information.

General: variability and model assumptions. Linear regression commonly relies on assumptions about the relation between predictors and outcome, residual behavior and variance. When variability is not adequately represented by the model, standard errors and confidence intervals may be unreliable. Over-dispersion is a general concern when observed variability exceeds what a model assumes; it can lead to uncertainty estimates that are too narrow if not addressed. The registry does not report evidence that over-dispersion was assessed or that it was present. It would be inappropriate to infer a violation or a correction from the record.

The record lists planning illustrations based on an observed effect, with a two-sided significance level of 0.05, equal allocation and a normal approximation. Such figures are planning illustrations, not observed power and not evidence of the trial’s original design justification. The record explicitly distinguishes them from the trial’s own design. No additional planning values or sample-size rationale are reported in the registry record.

10. Analyses Reported by the Sponsor

The sponsor-reported analysis is a linear regression comparison of final values, expressed as a mean difference of 0.52 with a 95% confidence interval from -0.18 to 1.2 and p = 0.14. The registry describes the analysis as a superiority comparison. The confidence interval includes zero, and the reported p-value does not meet the conventional 0.05 threshold. These statistics summarize the sponsor’s inferential analysis; they do not prove that the groups are equivalent or that no effect is possible.

The arm-level results report median change and inter-quartile range, which are different from the sponsor’s mean difference in final values. A simple recalculation from posted summaries would not reproduce the regression model’s result, because the model’s inputs, specification and underlying data are not reported. The registry does not post a separate recalculated estimate of the same regression contrast. The discrepancy flag is false, so the record does not identify a disagreement between the sponsor analysis and a recalculation. The sponsor analysis remains the reported inferential result, while the descriptive arm summaries give complementary information about change scores.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary PSQI outcome and secondary outcomes concerning mental and physical health and circadian rest-activity measures. It does not report a multiplicity adjustment, a hierarchy for testing outcomes, or a prespecified strategy for controlling the chance of false-positive findings across analyses. The number and organization of all planned analyses are not described in enough detail to determine how multiplicity was handled. Without an adjustment or clear testing hierarchy, p-values across multiple outcomes should not automatically be interpreted as if each were the only analysis.

Missing outcome data and analysis populations are also not described in the registry record. The posted arm-level summaries include participant counts, but the record does not explain whether these counts represent everyone randomized, everyone assessed at follow-up, or a particular analysis set. Reasons for missing PSQI measurements, the amount of missingness by group, and any imputation or sensitivity analyses are not reported in the registry record. The results therefore cannot establish whether missingness affected the comparison or how the analysis addressed it.

The registry does not report whether the primary analysis followed an intention-to-treat principle, whether any participants were excluded after randomization, or how deviations from the planned intervention were handled. These details matter because analysis population choices can affect the interpretation of a randomized comparison. The record also does not describe interim analyses, stopping rules or a sample-size justification. These elements should be treated as not reported in the registry record, rather than assumed to have been absent or present.

Overall, the registry contains a primary endpoint, a sponsor-reported regression comparison and arm-level summaries, but not enough information to audit the full analysis process. This limits assessment of selective reporting, multiplicity, missing-data handling and the exact population represented by the reported estimate.

12. 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 adverse-event record reports no serious adverse events in the indoor light intervention group among 98 at risk and none in the control group among 93 at risk. It also reports no other adverse events in either group, and no deaths in either group. No serious adverse event terms are listed.

These figures describe what was reported in the registry for the risk groups shown. They do not establish that no events occurred outside the reported collection or reporting process, and they do not quantify the probability of rare harms. The record does not provide details of how adverse events were defined, collected, monitored or attributed to the intervention. It also does not report event timing, severity grades beyond the serious versus other distinction, or the duration of safety follow-up.

Zero reported events should therefore be read as a finding about the registry’s posted safety counts, not as proof that the intervention is free of risk. The small number of reported events leaves limited information for comparing safety between groups, especially for uncommon outcomes. No causal conclusion about safety can be drawn from these counts alone.

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

The primary endpoint asks whether PSQI changed from baseline to the end of a 16-week light intervention period. The posted arm summaries show the same median change, 0.0, in both groups, with overlapping inter-quartile ranges. The sponsor’s reported regression analysis of final values estimates a mean difference of 0.52, with a 95% confidence interval from -0.18 to 1.2 and p = 0.14. Taken together, these statistics do not provide clear evidence of a difference in the reported comparison, while the interval leaves uncertainty about the direction and size of any difference.

This does not show that the groups are equivalent. Equivalence requires a prespecified margin and an analysis designed to test whether differences fall within that margin; those details are not reported in the registry record. Nor does a non-significant result demonstrate that the intervention has no effect. The confidence interval includes zero and also spans values on both sides, so the reported analysis is compatible with more than one interpretation of the underlying contrast.

The endpoint is a measure of sleep quality, not a complete assessment of health, wellbeing or lighting exposure. Although the broader project concerns indoor and outdoor lighting and several health-related domains, the primary result should not be extended to all those outcomes. Secondary outcomes are listed, but their results are not included in the registry record described for this primary comparison. The results also do not establish how the intervention was used in practice, how consistently participants used the lamps, or whether any effect persisted beyond the reported intervention period.

Finally, the summary statistics should be understood in light of the analysis mismatch: descriptive arm results concern median change, while the sponsor’s inferential result concerns a mean difference in final values. They are complementary, not directly equivalent. The registry provides limited information about missing data, model specification and analysis populations, so the precision and scope of conclusions are constrained by the reporting available.

14. Why This Trial Matters Statistically

This trial is statistically informative because it illustrates why the outcome definition and the reported contrast must be read together. The primary endpoint is framed as change from baseline, but the sponsor-reported regression parameter is a mean difference in final values. A reader who overlooks that distinction could incorrectly treat the arm summaries and regression estimate as two versions of the same calculation. They are not: the summaries describe different quantities and use different measures of center.

The record also highlights the difference between descriptive similarity and inferential evidence. The median change is 0.0 in each arm, but those values do not show that every participant had no change or that the complete distributions match. The inter-quartile ranges add information about spread, while the regression estimate and confidence interval address a modeled group contrast. Each contributes a different part of the statistical picture.

Another important lesson is how to read a non-significant result. The p-value of 0.14 and confidence interval that includes zero do not establish equivalence or absence of effect. The interval communicates that uncertainty remains, and the record does not report an equivalence margin or a design intended to establish equivalence. Statistical interpretation should distinguish lack of clear evidence for a difference from evidence that differences are negligible.

Finally, the record’s omissions matter. The analysis population, missing-data handling, model specification and multiplicity strategy are not reported. Those gaps do not invalidate the posted result, but they limit independent appraisal of how robustly it answers the stated question. The trial therefore offers a useful example of interpreting a reported effect while keeping the endpoint, analysis method, uncertainty and reporting limits in view.

15. 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 does not report eligibility criteria or detailed baseline characteristics, so the population represented by the results cannot be characterized fully. The title identifies older adults, but the primary outcome description does not specify age thresholds or other inclusion details. Generalizability beyond the enrolled participants cannot be assessed from the record alone.

Intervention delivery is described only as providing a lamp for home placement. Lamp specifications, exposure instructions, adherence, actual light exposure and co-interventions are not reported. Without these details, it is difficult to assess what intervention participants received in practice or how consistently it was implemented. The control group received no indoor light supplementation and assessment procedures only, but other aspects of usual conditions are not described.

The registry gives limited information about the statistical analysis. It reports linear regression and a mean difference in final values, but not the model specification, covariates, diagnostics or analysis population. The posted arm summaries use medians and inter-quartile ranges for change scores, a different estimand and summary approach. The lack of a reported analysis of the same change-score quantity makes direct comparison between the descriptive and inferential results difficult.

Missing outcome data, reasons for missingness, handling of incomplete observations, multiplicity procedures, interim analyses and sample-size justification are not reported in the registry record. The available adverse-event information contains no reported events, but details of collection and follow-up are also not reported. These limitations concern the reporting available; they should not be interpreted as evidence that a particular procedure was or was not used.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Population Based Lighting Study on Older Adults — NCT05676086.
• Registry record last updated 2026-05-26; status last verified 2025-02.
• Results first posted 2026-05-26.
• 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.

19. Record Summary

The registry describes a completed, randomized, parallel-group study with no masking, sponsored by Azienda Usl di Bologna. The study enrolled 191 participants and compares a home indoor light intervention with a control group receiving no indoor light supplementation and assessment procedures only.

The primary outcome is PSQI change from baseline to the end of the 16-week intervention period. Median change is 0.0 in both arms. The sponsor-reported linear regression analysis of final values gives a mean difference of 0.52, a 95% confidence interval from -0.18 to 1.2 and p = 0.14. Because the arm summaries and regression analysis describe different quantities, they should be interpreted separately.

The registry lists secondary health and circadian outcomes, but their results are not reported in the record. It also does not report key details about missing data, analysis populations, model specification or multiplicity handling. No serious adverse events, other adverse events or deaths are reported in either group.