Chronic Sleep Deprivation Among the Poor: A Lab-in-the-field Approach: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Chronic Sleep Deprivation Among the Poor: A Lab-in-the-field Approach (NCT03322358), a randomized na study. The primary outcome is Sleep Per 24 Hours.

2. Clinical Question

3. Trial Design

The registry record identifies this as a randomized, unmasked, factorial-model study, with the phase listed as NA. It is marked completed. The sponsor is the University of Pennsylvania. The recorded study start was 2017-10-23, and primary completion was 2019-12-31. The enrollment recorded in the registry is 452.

The six posted groups are control; naps only; home sleep aids only; home sleep aids with sleep incentives; naps with home sleep aids; and naps with home sleep aids and sleep incentives. This arrangement varies combinations of intervention components, but the registry record does not report a detailed factorial analysis plan or the intended contrasts between components. The arm descriptions say that interventions began after a baseline period. The control group continued normal sleep habits.

Sleep was measured with actigraphs worn consistently throughout the study. The registry describes daily measurement through study completion, which it characterizes as 28 business days, as well as intervening holidays and Sundays. Blood pressure was measured once in each block of three business days, with the measurement day randomized within the block. Depression and illness days were measured at baseline and endline according to their outcome descriptions.

The trial summary says the study aimed to assess sleep and health outcomes in a setting outside the sleep laboratories where much prior sleep research had taken place. The registry record does not report eligibility details, the recruitment process, the randomization procedure, allocation concealment, adherence, or follow-up completeness. A statistical analysis plan and the trial's original sample-size assumptions 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.

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.

Only the first 6 of 7 groups are shown.

6. Primary Endpoint

The registry lists the primary outcome as Sleep Per 24 Hours, measured over Actigraphs are worn consistently throughout the study. The sleep per 24 hour period outcome is then measured every day through study completion, which is 28 business days (as well as intervening holidays and Sundays) for each participant..

The primary endpoint is sleep per 24 hours, expressed in hours of sleep per 24-hour period. The registry defines it as the sum of night sleep and nap sleep, measured using actigraphs. The devices were worn consistently throughout the study, and the outcome was measured daily through completion, described as 28 business days plus intervening holidays and Sundays. The posted arm summaries report means and standard deviations.

Sleep duration is the direct outcome needed to examine whether the assigned sleep-intervention combinations were associated with differences in recorded sleep during the study. Including nap sleep in the daily total is relevant because an intervention offered a daytime nap opportunity: the endpoint captures that sleep as part of the full 24-hour measure rather than treating nighttime sleep as the only component. This is a general explanation of the outcome's measurement logic, not a claim about the investigators' analysis rationale, which is not reported in the registry record.

The arm means range from 5.61 hours in control to 6.35 hours in the group assigned naps, home sleep aids, and sleep incentives. The unadjusted mean difference for that group versus control is 0.74 hours, with a 95% confidence interval from 0.358 to 1.122 and p < 0.001. These summaries describe the posted outcome; they do not establish how the result would generalize beyond the study population or what changes in sleep mean for health.

7. Results Posted in the Registry

• Naps Only: mean 5.81 versus 5.61 on control; difference 0.20 (95% CI -0.19 to 0.59), standardized effect d = 0.16; z-test p = 0.316.
• Home Sleep Aids Only: mean 5.99 versus 5.61 on control; difference 0.38 (95% CI -0.01 to 0.77), standardized effect d = 0.31; z-test p = 0.055.
• Home Sleep Aids + Sleep Incentives: mean 6.23 versus 5.61 on control; difference 0.62 (95% CI 0.22 to 1.02), standardized effect d = 0.49; z-test p = 0.002.
• Naps + Home Sleep Aids: mean 6.08 versus 5.61 on control; difference 0.47 (95% CI 0.07 to 0.87), standardized effect d = 0.38; z-test p = 0.021.

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.

Blood Pressure

• 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 deviations.
• Result:
  • Naps Only: mean 0.14 versus 0.00 on control; difference 0.14 (95% CI -0.17 to 0.44), standardized effect d = 0.14; z-test p = 0.379.
  • Home Sleep Aids Only: mean 0.11 versus 0.00 on control; difference 0.11 (95% CI -0.20 to 0.42), standardized effect d = 0.11; z-test p = 0.494.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Naps Only 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 Home Sleep Aids Only and the control. The interval gives the range of effects that remain compatible with the data.

Depression

• 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 deviations.
• Result:
  • Naps Only: mean 4.68 versus 5.21 on control; difference -0.53 (95% CI -2.10 to 1.04), standardized effect d = -0.11; z-test p = 0.506.
  • Home Sleep Aids Only: mean 5.85 versus 5.21 on control; difference 0.64 (95% CI -1.05 to 2.34), standardized effect d = 0.12; z-test p = 0.458.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Naps Only 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 Home Sleep Aids Only and the control. The interval gives the range of effects that remain compatible with the data.

Number of Days of Illness in Past Week

• 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 deviations.
• Result:
  • Naps Only: mean 1.73 versus 2.08 on control; difference -0.35 (95% CI -1.03 to 0.34), standardized effect d = -0.16; z-test p = 0.321.
  • Home Sleep Aids Only: mean 1.97 versus 2.08 on control; difference -0.11 (95% CI -0.90 to 0.69), standardized effect d = -0.04; z-test p = 0.792.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Naps Only 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 Home Sleep Aids Only and the control. The interval gives the range of effects that remain compatible with the data.

The registry lists blood pressure, depression, illness days, inhibitory control, memory, attention, self-reported pain, and physical fitness as secondary outcomes. Numerical arm summaries and unadjusted comparisons are posted for blood pressure, depression, and illness days, but only for some intervention-versus-control contrasts. Numerical results for inhibitory control, memory, attention, self-reported pain, and physical fitness are not reported in the registry record. The listed results do not provide a complete set of estimates across every secondary outcome and intervention group.

These outcomes broaden the study beyond sleep duration, addressing health, symptoms, and cognitive measures named in the record. They relate to the primary result by raising the separate question of whether differences in recorded sleep coincided with differences in these other measures. They cannot be used to establish that a sleep difference caused a health or cognitive change. The numerical secondary summaries are not a substitute for the full outcome-specific analysis, and the registry does not report a unified analysis linking sleep changes to those endpoints.

In general, secondary findings warrant particular caution when many outcomes and comparisons are examined. If each is assessed separately without an adjustment strategy, the chance of at least one apparently notable result can increase. The record does not report a multiplicity plan, a hierarchy among secondary outcomes, or a complete analysis plan. Unadjusted secondary results therefore carry less weight than a fully prespecified, comprehensively reported analysis. This general concern does not establish that any specific posted secondary result is incorrect; it limits how strongly isolated results can be interpreted.

Since the primary endpoint measures sleep itself, the posted primary findings describe the most direct outcome of the assigned sleep interventions. The available secondary summaries are more limited and do not provide a complete basis for judging whether sleep differences corresponded to broader health changes.

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

• Blood Pressure
• Depression
• Number of Days of Illness in Past Week
• Inhibitory Control
• Memory
• Attention
• Self-reported Pain
• Physical Fitness

9. Statistical Interpretation

10. Statistical Methods Explained

The primary outcome is continuous: it is a numerical measure of sleep duration, and the registry reports a mean and standard deviation for each group. The comparison metric is a mean difference, calculated as intervention-group mean minus control-group mean. A positive value therefore indicates a higher posted mean sleep duration in the intervention arm than in control; it does not, by itself, identify why the means differ.

The registry record includes five intervention-versus-control mean differences, confidence intervals, and unadjusted p-values. These comparisons are described as recalculated, unadjusted summaries rather than as the sponsor's reported statistical analyses. No reported analyses are listed in the record. The recalculations use the posted arm means, standard deviations, and group sizes. The record does not describe a covariate-adjusted model, repeated-measures model, handling of daily measurements, or other detailed approach to account for how the sleep measurements were collected.

A confidence interval expresses the uncertainty around an estimated difference under the method used. The interval's endpoints are not guaranteed bounds on the true effect; rather, the interval is a statistical summary of the estimate's precision. An interval spanning zero is compatible with both a negative and a positive difference under the stated comparison. A p-value summarizes how unusual data at least as inconsistent with a zero difference would be under a specified null model. It is not the probability that the null hypothesis is true, and it does not measure the clinical importance of an estimate. These are general statistical explanations; the registry does not report a complete inferential-method description.

The registry classifies the design as randomized and factorial, and lists six groups corresponding to combinations of sleep interventions. Factorial designs can be used to examine individual intervention components and combinations, but the registry record does not report whether the posted analyses estimated component effects, interaction effects, or a factorial model. The displayed recalculated comparisons instead contrast each intervention arm with control. They should not be interpreted as direct tests of the differences between active arms or as estimates of isolated component effects.

Repeated observations are relevant to the primary endpoint because actigraph-based sleep was measured daily. An analysis that uses daily records may need to account for the fact that measurements from the same participant are related. A participant-level summary, a repeated-measures model, or another approach can address that structure in different ways. Which approach was used by the investigators is not reported in the registry record. The reported arm-level means and standard deviations do not reveal how within-person variation or incomplete daily recordings were handled.

For a continuous measure, the standard deviation describes the spread of individual observations around a group mean. It is different from the standard error, which describes uncertainty in an estimated mean. The posted sleep standard deviations indicate variation among participants' recorded sleep summaries; they do not directly tell the reader how accurately the group mean was estimated. The precision of a between-group estimate is better conveyed by its confidence interval.

General statistical context: over-dispersion occurs when observed variability is greater than a model assumes. If a model underestimates variability, standard errors can be too small and confidence intervals too narrow. This matters especially for outcomes measured repeatedly or as counts, but the registry record does not report that over-dispersion was assessed or that a particular model was used. It would be inappropriate to infer a specific model from the fact that the trial is called factorial.

The posted p-values are unadjusted. When several comparisons are examined, the chance of at least one apparently unusual result can increase if each is assessed separately at the same threshold. Adjustment procedures can address that issue, but the registry record does not say whether the investigators prespecified a primary contrast or adjusted secondary or multiple-arm comparisons. The posted results should therefore be considered within that reporting limitation. A complete prespecified analysis plan, test assumptions, and sensitivity analyses are not reported in the registry record.

Finally, the difference between a result being statistically distinguishable from zero under an unadjusted calculation and being important to participants is not resolved by a p-value alone. Interpretation also depends on measurement quality, study conduct, uncertainty, relevance of the endpoint, and the prespecified analysis. The registry record does not report a threshold for a meaningful difference in sleep duration, so no such threshold can be applied to these estimates.

11. Confidence Intervals and Precision

The confidence intervals for the primary mean differences convey how precisely each unadjusted comparison is estimated. For naps only, the estimate is 0.2 hours and the 95% interval runs from -0.191 to 0.591. For home sleep aids only, the estimate is 0.38 hours and the interval runs from -0.008 to 0.768. Both intervals include zero, so the posted data do not pin down the direction of the difference as precisely as an interval entirely on one side of zero would.

The intervals for home sleep aids with sleep incentives, naps with home sleep aids, and naps with home sleep aids and sleep incentives are 0.218 to 1.021, 0.071 to 0.869, and 0.358 to 1.122, respectively. Each has a positive lower endpoint. The interval for naps with home sleep aids is comparatively close to zero at its lower endpoint, indicating that the range of values compatible with that estimate includes relatively small positive differences. Across all arms, the intervals retain uncertainty about the size of the difference.

In general, a 95% confidence interval is produced by a procedure that, across repeated use under its assumptions, would cover the target value in 95% of such intervals. It is not a statement that there is a 95% probability that this particular interval contains the true value. The intervals accompany unadjusted recalculations, and the registry record does not report whether they account for multiple comparisons or repeated daily measurements.

Precision is influenced by the observed variability and group sizes as well as the estimated difference. The group sizes in the primary comparisons are between 74 and 77 participants per intervention arm, with 77 in control. The interval widths make clear that estimates are not exact, even when the interval excludes zero. These intervals concern mean sleep differences during the study as measured; they do not quantify uncertainty about longer-term outcomes, generalizability, or unreported endpoints.

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 participants or total events. They are illustrations based on the observed effect, not the original design assumptions for this trial; those original assumptions are not reported in the registry record. The listed planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, a normal approximation, and a target effect set to a stated fraction of the observed effect. The event-count convention for time-to-event outcomes does not apply to these continuous outcomes.

The 80% power column gives the illustrative number of participants per arm for a design with an 80% chance of detecting the specified effect under the planning assumptions, if that effect were the true effect and the assumptions held. The 90% power column gives the corresponding number for a design with a 90% chance. These are design-planning probabilities, not guarantees about the outcome of a future study. The figures are not observed power.

For each comparison, the planning illustration includes the full observed effect, 75% of the observed effect, and 50% of the observed effect. A smaller target effect requires more participants per arm to reach the same power under the stated assumptions. The values can differ greatly across comparisons because the observed effect sizes differ. A planning estimate based on the observed effect is particularly uncertain: the observed effect may not match the effect a new study would find, and the registry does not report a clinically meaningful target difference.

The sample-size estimates do not account for details that are not included in the stated assumptions, such as attrition, repeated-measure structure, multiplicity, or a factorial analysis. The figures are therefore illustrations of how sample size relates to effect size under a simplified two-arm planning calculation, not a reconstruction of how this completed study was designed.

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 = 0.200; pooled standard deviation σ = 1.229.
2. n = 2 × (1.96 + 0.842)² × 1.229² / 0.200² = 593.2, rounded up to 594 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 <- 0.199999999999999                      # observed difference in means
sigma <- 1.22946057006044                     # 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(3322358)
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 = 594 (80% power), 795 (90% power)
 75% of observed effect: n per arm = 1,055 (80% power), 1,412 (90% power)
 50% of observed effect: n per arm = 2,373 (80% power), 3,177 (90% power)
power.t.test: 594.2 
Simulated power at n = 594 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 record lists multiple intervention groups, several secondary outcomes, and repeated measurements for some outcomes. It does not report a prespecified multiplicity strategy, such as which comparisons were primary, whether a family of tests was defined, or whether any adjustment was applied. The posted intervention-versus-control p-values are unadjusted recalculations. Their interpretation should account for the fact that several contrasts are presented, rather than treating each p-value as if it were the only test conducted.

The registry record lists no reported analyses. It does not provide a complete statistical analysis plan, identify the primary analysis population, or say whether analyses followed intention-to-treat, per-protocol, or another population definition. Although randomization is recorded, the record does not explain how participants were analyzed if they did not receive or adhere to assigned interventions. The allocation sequence method and concealment procedures are also not reported in the registry record.

Missing-data handling is not reported. The registry does not state how many participants had complete actigraph records, whether daily measurements were missing, how missing endpoint values were treated, or whether any sensitivity analyses were performed. It also does not describe imputation or the assumptions made about missingness. The posted arm sizes alone do not establish that all randomized participants contributed complete data to each outcome.

Repeated daily sleep observations can require methods that account for within-participant dependence, while the registry presents arm-level summaries. Whether the investigators analyzed daily records, participant-level averages, or another summary is not reported. The registry also does not report model diagnostics, covariate adjustment, or handling of deviations from the assigned intervention. These omissions limit the ability to reproduce or fully evaluate the analyses from the record.

For secondary endpoints, the record names more outcomes than it numerically reports. A complete outcome-by-arm account, including analysis populations, missingness, and prespecified testing procedures, is not reported in the registry record. This limits assessment of selective reporting and the strength of inferences across the full set of planned outcomes.

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's adverse-event summaries list six groups: control, naps only, home sleep aids only, home sleep aids with sleep incentives, naps with home sleep aids, and naps with home sleep aids and sleep incentives. For each group, the number at risk for serious adverse events and other adverse events matches the posted group size: 77, 75, 75, 74, 75, and 76, respectively. The recorded counts affected by serious adverse events and other adverse events are 0 in every group. The number of deaths is also 0 in every group.

These figures describe what was recorded in the registry; they are not a statistical comparison of safety between groups. Zero recorded events does not prove that no event occurred, and the record does not describe how events were solicited, defined, monitored, or adjudicated. No serious-event terms are listed. The available safety summary should therefore be read as a registry report of zero affected participants in the listed groups, not as a complete evaluation of all possible harms.

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

The posted primary outcome is objective sleep duration measured using actigraphs. Across the intervention groups, the arm means were numerically higher than the control mean, with the largest posted mean in the group assigned naps, home sleep aids, and sleep incentives. The unadjusted estimates and confidence intervals indicate that the precision differs across comparisons: some intervals include zero, while others have positive lower endpoints. Together, these results describe the posted group summaries and their uncertainty; they do not alone establish the broader consequences of the interventions.

The endpoint is not a health outcome. Although the study summary describes a focus on cardiovascular health and cognitive outcomes, a difference in sleep duration cannot by itself show a corresponding change in blood pressure, depression, cognition, pain, fitness, or illness. Numerical secondary results are limited in the registry record, and results for several listed secondary outcomes are not reported there. No conclusion about those outcomes can be drawn from the primary sleep comparison alone.

The record does not report a threshold for a meaningful change in sleep duration. Statistical evidence against a zero difference and practical importance are distinct questions. The unadjusted p-values also do not account for a reported multiplicity plan, and such a plan is not reported. The p-values should not be treated as definitive evidence that any one intervention combination is superior to another; the posted comparisons are each against control, not direct active-arm comparisons.

Random assignment is recorded, but important details of implementation and analysis are absent, including adherence, missing-data procedures, the analysis population, and how daily measurements were handled. Those gaps affect how confidently the posted summaries can be interpreted. The findings describe the participants and measurement period represented in the registry record; they do not establish persistence after study completion or generalizability to other populations and settings.

Accordingly, the result is best understood as evidence about measured sleep duration in this completed randomized study, with estimates that have varying precision and reporting limitations. It is not a standalone answer about health benefit, long-term outcomes, or the relative effects of the different intervention components.

17. Why This Trial Matters Statistically

This trial is statistically notable for pairing a field-based randomized intervention study with objective actigraph measurement of sleep. The registry describes daily sleep measurement in low-income adults in Chennai, India, and intervention combinations involving nap opportunities, home sleep aids, and incentives. Its setting and measurement approach address a different research context from controlled sleep laboratories, as the study summary itself explains.

The factorial label and multiple intervention combinations also make careful contrast definition important. A comparison of each active group with control is not the same as estimating the contribution of a single component or testing whether components interact. The registry record does not report those component-level analyses, so the posted results should be read as specific arm-versus-control comparisons rather than as a full factorial assessment.

The continuous primary endpoint supports interpretation through means, standard deviations, mean differences, and confidence intervals. Looking at uncertainty alongside point estimates is particularly useful when group sizes are in the range of 74 to 77 participants per intervention arm and when several comparisons are presented. The confidence intervals show that the estimates vary in precision and should not be reduced to a binary reading of p-values.

The record also illustrates why design and reporting details matter. Daily measurements, several intervention arms, and multiple secondary outcomes raise questions about repeated observations, multiplicity, missing data, and analysis populations. Those methods are not fully documented in the registry record. This is not a judgment that the analyses were inadequate; it identifies information needed to assess the strength and reproducibility of the posted results.

Overall, the study offers a useful example of interpreting an objective behavioral endpoint in a randomized field setting while distinguishing observed arm summaries from broader health conclusions.

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 does not give detailed eligibility criteria, participant baseline characteristics, recruitment methods, or the randomization and allocation-concealment procedures. Although it identifies a low-income adult population in Chennai, those details would help assess who the findings represent and how groups were formed.

The study is described as randomized, unmasked, and factorial, but the record does not report adherence, intervention uptake, contamination between groups, or whether all assigned participants were included in the analyses. Some intervention descriptions are truncated, limiting the available detail about implementation. The record also does not report how actigraph non-wear or missing daily measurements were addressed.

Daily actigraph measures create a repeated-observation structure, but the analysis approach for those measurements is not reported. The posted summaries do not indicate whether the investigators used daily readings or participant-level summaries, or how within-person dependence was handled. The trial's original sample-size justification and its statistical analysis plan are not reported, including the intended factorial contrasts and multiplicity strategy.

Numerical results are incomplete across secondary outcomes and intervention comparisons. This limits interpretation of the broader health and cognitive aims described in the study summary. The adverse-event record reports zero affected participants in the listed categories but does not describe safety-monitoring or event-collection methods. Finally, the registry results address the study's measured period; longer-term follow-up is not reported in the registry record.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Chronic Sleep Deprivation Among the Poor: A Lab-in-the-field Approach — NCT03322358.
• Registry record last updated 2023-09-21; status last verified 2022-11.
• Results first posted 2023-09-21.
• 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

NCT03322358 is a completed randomized, unmasked study listed with a factorial model and phase NA. The sponsor is the University of Pennsylvania; recorded start and primary-completion dates are 2017-10-23 and 2019-12-31, and enrollment is 452. The study summary describes low-income adults in Chennai, India.

The primary endpoint is actigraph-measured sleep per 24 hours, combining night and nap sleep. The posted group summaries show mean sleep ranging from 5.61 hours in control to 6.35 hours in the group assigned naps, home sleep aids, and sleep incentives. The corresponding unadjusted mean difference is 0.74 hours (95% CI 0.358 to 1.122; p < 0.001). The record lists several health and cognitive secondary outcomes, but numerical reporting is limited. Analysis-plan, missing-data, and multiplicity details are not reported in the registry record.