PATient Navigator to rEduce Readmissions: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for PATient Navigator to rEduce Readmissions (NCT02114515), a randomized na study. The primary outcome is PROMIS Emotional Distress-Anxiety (v1.0, SF4a).

2. Clinical Question

The central statistical question concerns the difference between the groups in mean change, not whether every participant’s anxiety score changed in the same way. The recorded mean change was more negative in the PArTNER group than in the usual-care group. The unadjusted mean difference was -1.5 T-score points, with a confidence interval from -2.886 to -0.114 and an unadjusted p-value of 0.034. In the registry’s direction convention, a negative change represents improvement, so the group-level estimate points toward a larger average decrease in the intervention group. The estimate and its uncertainty describe the posted data; they do not establish what happened for each individual.

The question is limited to this measured outcome and follow-up time. It does not by itself answer whether the intervention changed readmissions, other health outcomes, or patient experience. Although the trial’s title and summary concern readmissions and hospital-to-home transition, the primary outcome listed in the registry is anxiety-related T-score change. The registry record does not report a primary readmission result in the posted outcome data.

3. Trial Design

The registry identifies this as a randomized, single-masked, parallel-group trial and lists the study status as completed. Enrollment was 1029. The sponsor was the University of Illinois Chicago. The registry records a study start of 2014-07 and primary completion in 2016-10. The phase is listed as NA. The trial included people hospitalized with chronic obstructive pulmonary disease, heart failure, sickle cell disease, myocardial infarction, or pneumonia.

The comparison was between hospital usual care and usual care plus PArTNER. Usual care consisted of written discharge instructions before discharge. The intervention arm is described as adding a community health worker and a peer-led telephone support line to usual care. The summary frames the program as support for the transition from hospital to home. The registry describes the community health worker as providing social support and literacy-appropriate education, but the full intervention description is not reported in the registry record. The registry identifies the study as single-masked but does not report who was masked.

The primary outcome was PROMIS Emotional Distress-Anxiety (v1.0, SF4a), expressed as a T-score. It was assessed as change from baseline to 30 days post discharge, defined as the 30-day score minus the baseline score. The primary arm summaries include 406.0 participants in usual care and 391.0 in usual care plus PArTNER. The registry warns that missing data can make the number analyzed smaller than the numbers in participant flow. These outcome-specific counts should not be treated as the total number enrolled or as proof that every enrolled participant contributed a primary outcome.

Other outcomes recorded include PROMIS emotional support, instrumental support, global physical health, global mental health, anxiety, and informational support. The registry lists 18 posted secondary outcomes. It does not report the trial’s original sample-size justification, a detailed statistical analysis plan, the allocation procedure beyond randomized assignment, the identity of the masked party, or how missing outcome data were handled. Those omissions limit how fully the posted results can be connected to the trial’s planned analysis.

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:

Analysis population for the primary outcome, as defined in the registry: Due to missing data, the number of participants analyzed may be less than the numbers provided in the Participant Flow module.

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 PROMIS Emotional Distress-Anxiety (v1.0, SF4a), measured over 30 days post discharge.

The primary endpoint is change in PROMIS Emotional Distress-Anxiety T-score from baseline to 30 days post discharge. The change is calculated as the score at 30 days minus the baseline score. The registry defines a change below 0 as improvement, a change equal to 0 as no change, and a change above 0 as worsening. This explicit direction matters: a more negative change represents a larger decrease in the measured anxiety score.

A continuous score retains information about the size of changes across the scale rather than reducing each participant to a category such as improved or not improved. The reported group summaries are means, so they describe average change in each arm. They do not show how changes were distributed among participants, whether most participants moved in the same direction, or whether any particular change was important to an individual. The registry identifies the measure as a T-score but does not report a threshold for a clinically meaningful difference.

The 30-day assessment makes the endpoint a short-term measure after discharge. It is relevant to a program focused on the transition home, but it is not itself a direct measure of readmission, length of stay, or the full patient experience. The registry’s outcome description supplies the measurement and time point, but does not explain the rationale for selecting this endpoint over other possible outcomes. That rationale is not reported in the registry record.

7. Results Posted in the Registry

• Usual Care + PArTNER: mean -1.70 versus -0.20 on control; difference -1.50 (95% CI -2.89 to -0.11), standardized effect d = -0.15; 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.

PROMIS Emotional Support (v2.0, SF4a)

• 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:
  • Usual Care + PArTNER: mean -0.10 versus -0.10 on control; difference 0.00 (95% CI -1.39 to 1.39), standardized effect d = 0.00; z-test p = 1.000.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Usual Care + PArTNER and the control. The interval gives the range of effects that remain compatible with the data.

PROMIS Instrumental Support (v2.0, SF4a)

• 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:
  • Usual Care + PArTNER: mean 0.60 versus 1.00 on control; difference -0.40 (95% CI -1.79 to 0.99), standardized effect d = -0.04; z-test p = 0.572.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Usual Care + PArTNER and the control. The interval gives the range of effects that remain compatible with the data.

PROMIS Global Health, Physical (v1.1, SF)

• 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:
  • Usual Care + PArTNER: mean 4.40 versus 4.60 on control; difference -0.20 (95% CI -1.31 to 0.91), standardized effect d = -0.03; z-test p = 0.724.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Usual Care + PArTNER and the control. The interval gives the range of effects that remain compatible with the data.

The posted secondary results cover related PROMIS measures of emotional support, instrumental support, global physical health, and other domains. Across the secondary comparisons with numerical results available, estimates are generally near zero relative to their uncertainty, and the reported intervals include no difference. This overall pattern does not present a clear, consistent separation between the groups across the measured secondary domains.

The primary anxiety result is more distinct in its direction and unadjusted statistical evidence than this broad secondary pattern. That contrast does not make the secondary outcomes irrelevant: they help describe whether the primary finding appears alongside similar differences in other reported dimensions of health and support. But measures of emotional support, instrumental support, or global health are not interchangeable with anxiety, and results on one do not validate or invalidate another automatically.

The registry lists 18 posted secondary outcomes, while the available information does not establish a complete multiplicity-adjusted analysis. When many outcomes are examined, an unadjusted p-value for any one result does not account for the number of questions considered. The chance of observing a seemingly notable result somewhere in a broad set can increase even in the absence of a systematic effect. For that reason, secondary comparisons generally carry less confirmatory weight unless they were prespecified and interpreted with a stated approach to multiplicity.

The record lists no reported analyses for the secondary outcomes. It does not explain which outcomes were prioritized, whether any were exploratory, or whether adjustments were made outside the posted information. The safest reading is descriptive: the secondary results provide context, but they do not amount to a coordinated set of confirmatory tests. The primary result should likewise be understood in the context of the full outcome family, rather than treated as wholly independent of the other questions asked.

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

• PROMIS Emotional Support (v2.0, SF4a)
• PROMIS Instrumental Support (v2.0, SF4a)
• PROMIS Global Health, Physical (v1.1, SF)
• PROMIS Global Health, Mental (v1.1, SF)
• PROMIS Emotional Distress-Anxiety (v1.0, SF4a)
• PROMIS Informational Support (v2.0, SF4a)
• PROMIS Emotional Support (v2.0, SF4a)
• PROMIS Instrumental Support (v2.0, SF4a)

9. Statistical Interpretation

10. Statistical Methods Explained

What does a mean change represent?

For each participant with a recorded score at both time points, change is the 30-day post-discharge T-score minus the baseline T-score. A negative value means the later score is lower than baseline, according to the registry’s stated convention. The arm-level means then average these participant-level changes. A mean is useful for comparing groups on a continuous measure, but it can conceal variation: participants may have had quite different individual changes even when the average is similar. The arm summaries do not provide the full distribution of changes.

The primary summaries report means and standard errors. A standard error describes the precision of an estimated mean, rather than the spread of individual participants’ scores. The reported standard error is 0.5 in each primary arm. It should not be mistaken for a standard deviation. The recalculated standardized effect uses the pooled standard deviation of 9.983, which is a measure of participant-level variation estimated from the posted summaries. These quantities answer different questions: a standard error concerns uncertainty in a mean, while a standard deviation concerns variability among observations.

How should the group difference be read?

The mean difference is the intervention-arm mean minus the usual-care mean. The posted primary arm means are -1.7 for usual care plus PArTNER and -0.2 for usual care; the reported difference is -1.5. The negative sign is meaningful because the registry defines negative changes as improvement. It indicates that the intervention arm’s average change was more negative. A difference in means does not identify the cause of that difference, and a group average does not describe every participant’s outcome.

The unadjusted calculation is a direct comparison of the posted summary statistics, without adjustment for other variables. The recalculated standardized difference, -0.150, expresses the difference relative to the pooled standard deviation. Standardization can help compare the magnitude of effects measured on different scales, but it is not a replacement for understanding the original T-score scale. The registry does not give a clinical-importance threshold for the primary measure.

What does the p-value contribute?

The unadjusted p-value for the primary comparison is 0.034. In general, a p-value describes how compatible the observed data are with a specified null model of no group difference, under the assumptions used for the test. It is not the probability that the null hypothesis is true, nor does it quantify the probability that the result occurred by chance in a literal sense. It also does not measure the size or clinical importance of the difference. The registry does not report a detailed primary analysis method, so the precise model and assumptions behind any sponsor analysis are not reported in the registry record.

The primary confidence interval, from -2.886 to -0.114, gives a range of values compatible with the observed data under the interval procedure. In general, a confidence interval communicates both the direction and precision of an estimate. A narrower interval indicates greater precision than a wider interval for the same measure and analysis. The interval does not account for every possible source of uncertainty, including unreported choices in outcome handling or analysis.

Why can over-dispersion matter?

General statistical explanation: over-dispersion occurs when observed variability is greater than a model assumes. If a model underestimates variability, standard errors can be too small and uncertainty intervals can be too narrow. This concern is especially relevant for count outcomes or clustered data, where observations may be more alike within groups than a simple model allows. The primary endpoint is a continuous score, and the registry does not report an over-dispersion analysis or a model-based adjustment for it. No conclusion about over-dispersion in this trial can be drawn from the posted summaries alone.

What is known about the analysis population?

The number analyzed for the primary measure is 406.0 in usual care and 391.0 in usual care plus PArTNER. The registry notes that missing data may result in fewer participants being analyzed than appear in participant flow. It does not report the amount or pattern of missingness, the method used to address missing values, or whether the primary analysis followed an intention-to-treat principle. Those details matter because a complete-case comparison can differ from an analysis that includes all randomized participants under explicit assumptions about missing outcomes.

The registry records zero reported analyses in its analysis listing. It nevertheless posts arm-level outcome summaries and an unadjusted comparison. A detailed protocol-defined model, covariate adjustment, test specification, or analysis population definition is not reported in the registry record. The recalculated comparison should therefore be read as an unadjusted analysis of the posted summaries, not as a reconstruction of an undocumented sponsor model.

How do the secondary measures fit?

The secondary outcomes are also continuous PROMIS measures, with means and standard errors recorded at 30 days post discharge. Their comparisons are unadjusted. In general, examining several outcomes increases the chance that at least one result may appear unusual even if no underlying effect exists across the set. Multiplicity methods address that issue by controlling the interpretation of multiple tests, but no such method is reported in the registry record. Secondary results can describe whether the primary pattern is echoed across related measures, but they should not be treated as independent confirmation without a stated analysis plan.

The planning calculations use a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. They illustrate required participants per arm for a specified fraction of the observed effect, rather than documenting the assumptions used to design this trial. Their purpose is to show how sample-size requirements change when the target effect changes. They do not indicate the probability that this completed trial detected an effect.

11. Confidence Intervals and Precision

The primary mean difference is -1.5 T-score points, with a confidence interval from -2.886 to -0.114. The interval lies entirely on the negative side of zero. In the registry’s scoring convention, this direction corresponds to a more negative average change in the intervention group. The distance between the interval endpoints indicates that the data remain compatible with a range of possible mean differences, from a value close to zero to a more pronounced negative difference. The point estimate is not known with exactness.

In general, a confidence interval is a way to express uncertainty around an estimate. Its endpoints depend on the observed data and the statistical procedure. If the interval is narrow, the estimate is more precise; if it is wide, a broader range of values remains compatible with the data. The interval does not guarantee that the true effect lies within its endpoints, and it does not capture uncertainty arising from unreported design or analysis choices. The confidence level for the posted interval is not reported in the registry record, so it should not be assigned a level that the record does not state.

The unadjusted p-value is 0.034. Read together, the p-value and interval indicate that the posted primary comparison is separated from no difference under the calculation used, while the interval still allows a range of effect sizes. Statistical evidence against a zero difference does not establish clinical importance. The registry does not report a threshold for meaningful change in this PROMIS anxiety measure, and the interval alone cannot supply one.

Precision is also affected by who contributed outcome data. The registry gives 406.0 and 391.0 as the primary analyzed group counts and notes that missing data may make analyzed numbers smaller than those in participant flow. The handling of missing data is not reported in the registry record. Consequently, the interval reflects the uncertainty represented by the posted comparison, but it cannot resolve whether missing outcome data could have influenced the estimate.

The standardized difference of -0.150 describes the mean difference in relation to the pooled variability. It is a supplementary way to express magnitude, not an interval and not a clinical threshold. Neither standardization nor a p-value removes the need to consider the scale, follow-up time, analysis population, and incomplete reporting when judging the precision and meaning of the result.

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 use a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. For each planning scenario, the assumed effect is a stated fraction of the observed effect. These are planning illustrations based on the observed comparison, not the original sample-size assumptions for the trial; the trial’s own design assumptions are not reported in the registry record.

The 80% power column gives the illustrative number of participants per arm needed, under those assumptions, for an analysis to have an 80% chance of detecting the selected effect if that effect were the true one. The 90% power column uses a higher target chance of detection and therefore generally requires more participants per arm. These are design-planning quantities; this is not observed power.

The primary comparison’s planning figures are 696 participants per arm for 80% power and 931 for 90% power when planning for 100% of the observed effect. For 75% of the observed effect, the corresponding figures are 1,237 and 1,655. For 50% of the observed effect, they are 2,782 and 3,724. Smaller target effects require more participants because they are harder to distinguish from ordinary sampling variation.

These figures should not be read as evidence that the completed trial was adequately or inadequately designed. They use the observed effect as their starting point, and the observed effect itself is uncertain. A planning calculation also depends on assumptions about the analysis and outcome variability. The stated assumptions make the illustration interpretable, but they do not replace the original protocol’s design rationale, which is not reported in the registry record.

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 = -1.500; pooled standard deviation σ = 9.983.
2. n = 2 × (1.96 + 0.842)² × 9.983² / -1.500² = 695.3, rounded up to 696 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 <- -1.5                      # observed difference in means
sigma <- 9.98300442544681                     # 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(2114515)
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 = 696 (80% power), 931 (90% power)
 75% of observed effect: n per arm = 1,237 (80% power), 1,655 (90% power)
 50% of observed effect: n per arm = 2,782 (80% power), 3,724 (90% power)
power.t.test: 696.3 
Simulated power at n = 696 per arm: 0.800 (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 lists 18 posted secondary outcomes in addition to the primary endpoint. Testing or estimating many outcomes creates a multiplicity issue: even when there is no true group difference for any of them, a set of unadjusted tests can include a result that appears more compelling than expected. This is a general statistical concern, not evidence that any particular result in this trial is false.

The posted primary p-value is unadjusted, and the secondary comparisons are also presented as unadjusted. A multiplicity correction, hierarchical testing strategy, or other prespecified method is not reported in the registry record. Nor does the record identify a hierarchy among secondary outcomes or state which outcomes were intended as confirmatory. Without that context, the primary result has to be interpreted alongside the number of other reported questions, and secondary p-values should be treated as descriptive rather than definitive evidence.

Missing data are another important limitation. The registry explicitly notes that the number analyzed may be smaller than the participant-flow numbers because of missing data. The primary analyzed counts are 406.0 in usual care and 391.0 in usual care plus PArTNER. The record does not report how many outcome values were missing, reasons for missingness, whether missingness differed by arm, or the method used to handle missing values. It also does not report sensitivity analyses examining alternative assumptions about missing outcomes.

The registry does not state whether the analysis followed an intention-to-treat principle, how participants were assigned to the analysis population, whether covariates were included, or what statistical model was used for the sponsor’s primary analysis. The analysis listing contains no reported analyses. These items are not reported in the registry record; their absence should not be filled in by assuming common practice. They matter because choices about population, missingness, and adjustment can affect estimates and uncertainty.

The original sample-size rationale, interim-analysis plan, and stopping rules are also not reported in the registry record. The planning illustrations based on the observed effect answer a different question from the trial’s original design rationale. Taken together, these reporting gaps limit assessment of prespecification, robustness, and the role of multiplicity. They do not erase the posted numerical results, but they narrow the conclusions that can be drawn from them.

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 5% reporting threshold.

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: all-cause death (2.3%).

The registry records serious adverse events among 511 participants at risk in usual care and 518 at risk in usual care plus PArTNER. It lists 12 participants affected in each group. The registry’s separate death counts are 12 in each group, and all-cause death is listed as a serious adverse event term at 2.3%. The record also lists zero other adverse events in the at-risk groups.

These figures are descriptive counts, not a complete comparison of event timing, severity, relatedness, or clinical course. The registry does not provide a detailed breakdown of the serious events beyond the listed term, nor does it report a formal statistical comparison of adverse-event rates. The at-risk counts for safety are not the same as the primary outcome analysis counts, so the populations should not be conflated.

Equal recorded counts do not establish that the groups had identical safety experience, particularly without details about event definitions, follow-up, and analysis. Conversely, these counts alone do not establish a difference between groups. The registry information supports reporting what was recorded, but is not enough to infer causation or make a complete assessment of safety.

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

The posted primary data show that mean anxiety T-score change from baseline to 30 days post discharge was more negative in the usual-care-plus-PArTNER group than in the usual-care group. The estimated difference was -1.5 T-score points, and the confidence interval from -2.886 to -0.114 excludes zero under the posted calculation. The unadjusted p-value was 0.034. In the registry’s definition, a negative change indicates improvement, so the observed direction is consistent with a larger average decrease in anxiety score in the intervention arm.

This finding does not establish that the intervention caused the observed difference. A randomized design supports a comparative interpretation, but understanding the result also requires details of allocation, masking, outcome handling, analysis populations, and missing data. Several of those details are not reported in the registry record. The result also does not mean that all participants experienced a decrease, or that the average difference was important to every participant.

The endpoint is a PROMIS anxiety T-score change, not a direct measure of readmission. The study summary describes a program intended to improve the hospital-to-home transition, but the primary outcome result does not show whether readmissions changed. Nor does it establish an effect on the other clinical conditions listed in the registry, on long-term outcomes, or on outcomes beyond the recorded follow-up.

The secondary findings do not show a similarly clear pattern of group differences across the posted measures, and no multiplicity adjustment is reported. The primary result should therefore be considered alongside the broader set of outcomes and the limitations of the record. The registry does not report a threshold for clinically meaningful change, so statistical evidence for a difference should not be equated with evidence of a meaningful benefit. The record supports a careful account of the measured score and its uncertainty, not medical advice or a broad conclusion about the intervention.

17. Why This Trial Matters Statistically

This trial illustrates why statistical interpretation begins with the endpoint actually recorded, rather than the broad motivation described in a study summary. Although the program was developed around the hospital-to-home transition and readmissions, the posted primary result concerns change in PROMIS anxiety T-score at 30 days post discharge. That distinction prevents a result on one outcome from being misrepresented as evidence about another.

It also shows the value of interpreting an estimate together with its confidence interval and scale. The primary estimate points toward a more negative average change in the intervention arm, and its interval lies on the same side of zero. Yet the interval permits a range of magnitudes, and the registry gives no clinical-importance threshold. The p-value alone cannot answer whether the difference is consequential to patients.

Multiple outcomes and incomplete reporting are important statistical context. The registry lists many secondary outcomes, but does not report a multiplicity strategy, a detailed analysis plan, or the method used to handle missing outcomes. These omissions affect how strongly a result can be treated as confirmatory and how confidently the analysis population can be understood. They also demonstrate why a statistically notable primary comparison should not be interpreted in isolation from the full set of measured outcomes.

Finally, the planning calculations based on the observed effect are illustrations for future planning, not evidence about the trial’s original design or its achieved power. Keeping planning quantities separate from observed findings helps avoid a common interpretive error. In combination, these features make the trial useful for learning how endpoint definition, uncertainty, missing data, multiplicity, and planning assumptions shape what can responsibly be concluded from a registry result.

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 important design and analysis details unspecified. It identifies randomization and single masking, but does not state who was masked or give further detail about how random allocation was implemented. The intervention description is incomplete, and the record does not provide enough detail to reproduce all components of the community health worker and peer-led support program.

The primary outcome analysis includes 406.0 participants in usual care and 391.0 in usual care plus PArTNER, while the registry cautions that missing data may reduce analyzed numbers relative to participant flow. The amount, pattern, and handling of missing data are not reported. The record also does not state whether the analysis included all randomized participants, whether covariate adjustment was used, or how the primary statistical test was specified.

The trial includes several different conditions, but the posted primary summary does not provide condition-specific outcome comparisons. It therefore cannot show whether the overall estimate was similar across those clinical groups. The follow-up for the primary endpoint is 30 days post discharge, so the result does not describe persistence beyond that assessment. The primary measure is a symptom-related score, not a direct readmission outcome, despite the study’s broader readmission focus.

The registry does not report a clinically meaningful threshold for the PROMIS anxiety score, the original sample-size justification, interim analyses, a multiplicity plan, or sensitivity analyses. It lists numerous secondary outcomes without a reported analysis hierarchy. The safety data are presented as counts and do not include a detailed event breakdown or a formal comparison. These reporting limitations do not change the posted estimates, but they constrain interpretation of causality, clinical importance, generalizability across conditions, and robustness to analytical choices.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: PATient Navigator to rEduce Readmissions — NCT02114515.
• Registry record last updated 2019-02-15; status last verified 2018-10.
• Results first posted 2019-02-15.
• 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

NCT02114515, titled PATient Navigator to rEduce Readmissions, is recorded as a completed randomized, single-masked, parallel-group study sponsored by the University of Illinois Chicago. The registry lists enrollment of 1029 and study dates of 2014-07 to primary completion in 2016-10. The population includes people hospitalized with chronic obstructive pulmonary disease, heart failure, sickle cell disease, myocardial infarction, or pneumonia.

The comparison was hospital usual care—written discharge instructions—with usual care plus a community health worker and peer-led telephone support line. The primary endpoint was change in PROMIS Emotional Distress-Anxiety T-score from baseline to 30 days post discharge. The posted mean difference was -1.5, with a confidence interval from -2.886 to -0.114 and an unadjusted p-value of 0.034. The registry defines negative change as improvement.

The result is an unadjusted comparison of the posted summaries. Missing-data handling, the original design assumptions, the detailed analysis plan, and multiplicity procedures are not reported in the registry record. The findings describe the measured outcome and should not be taken as a direct readmission result or as proof of clinical importance.