Basel Discharge Communication Project: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Basel Discharge Communication Project (NCT01540266), a randomized na study. The primary outcome is Number of Items Recalled From the Communication in the Video.

2. Clinical Question

The central scientific question concerns whether the organization of a spoken discharge message is associated with immediate recall in this video-based task. Because the comparison groups also differ by student group, comparisons against the reference group do not isolate a single communication-format contrast across all students. The registry record does not report a single pooled comparison of structured versus non-structured communication across student groups.

3. Trial Design

The Basel Discharge Communication Project, identified as NCT01540266, is recorded as a completed randomized, parallel-group study with single masking. Its phase is listed as not applicable. Enrollment is 242. The registry record gives a start date of 2011-07 and primary completion in 2013-10. The sponsor is University Hospital, Basel, Switzerland.

The study used videos in which the same physician conveyed identical information to an older patient played by an actor. The communication was presented in structured or non-structured form. The six reported groups combine the communication condition with three student categories: first-year psychology students, first-year medical students and third-year medical students. The registry record does not describe the randomization procedure, allocation ratio, concealment process, or the party who was masked. Although masking is recorded as single, who was masked is not reported in the registry record.

The primary outcome was the number of items recalled from the video at 5 minutes. Each video conveyed 28 information items. Participants' recall protocols were evaluated by two independent raters; one rated all protocols and the other rated a subset. In case of disagreement, the raters reached consensus through joint analysis and discussion. The registry description reports a Cohen's kappa of 0.74 for agreement between raters.

Two secondary outcomes are listed: estimates of the quality of the communication and estimates of the current status. Their detailed definitions, results, and timing are not reported in the registry record. The registry record does not report the planned sample-size justification, original statistical analysis plan, or the intended handling of the multiple student groups and outcomes.

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 Number of Items Recalled From the Communication in the Video, measured over 5 minutes.

The primary endpoint is the mean number of information items recalled from the communication in the video, measured at 5 minutes. It is a continuous count-like measure bounded by the 28 items conveyed in the video. The registry describes the content as identical in the structured and non-structured videos, so the outcome focuses on immediate recall rather than the amount of information delivered.

Recall protocols were evaluated by two independent raters. One rater scored every protocol, while the other assessed a subset. Their agreement is reported as Cohen's kappa of 0.74, which the registry description characterizes as substantial interrater reliability according to its cited convention. Disagreements were resolved by joint review and discussion. This process is relevant because judgments about whether a response recalls an item can involve rating decisions; consensus helps define the final recorded score, although the record does not describe the scoring rubric in further detail.

Means describe the average number of items recalled in each group, while standard deviations describe how much individual scores varied around those averages. The endpoint is useful for comparing immediate memory performance across the study's video conditions and student groups. It does not directly measure longer-term recall, understanding, communication quality, or what happens when patients receive discharge information in routine care. The registry record does not report a minimally important difference for this outcome.

7. Results Posted in the Registry

• First Year psychology_non-structured: mean 5.71 versus 8.12 on control; difference -2.41 (95% CI -4.02 to -0.80), standardized effect d = -0.59; z-test p = 0.003.
• First Year medical_structured: mean 9.95 versus 8.12 on control; difference 1.83 (95% CI 0.13 to 3.53), standardized effect d = 0.39; z-test p = 0.035.
• First Year mediclal_non-structured: mean 8.76 versus 8.12 on control; difference 0.64 (95% CI -1.17 to 2.45), standardized effect d = 0.15; z-test p = 0.489.
• Third Year medical_structured: mean 13.45 versus 8.12 on control; difference 5.33 (95% CI 3.07 to 7.59), standardized effect d = 1.22; z-test p = <0.001.

8. Statistical Interpretation

9. Statistical Methods Explained

The primary outcome is continuous in the statistical sense used for reporting: each participant receives a numerical recall score, and the registry summarizes each group with a mean and standard deviation. The mean is the arithmetic average of scores in a group. The standard deviation describes the spread of scores around that average. A larger standard deviation means individual scores were more dispersed, not that the group mean is less valid by itself.

What does a mean difference represent?

A mean difference compares the average score in one group with the average score in another. The posted comparisons use first-year psychology students in the structured condition as the reference. For groups shown the non-structured video, the estimate is expressed as that group's mean minus the reference mean; for other groups, it likewise represents the contrast with the reference group. A negative estimate indicates a lower observed group mean than the reference, while a positive estimate indicates a higher one. These contrasts are not all the same scientific comparison: some change communication condition, some change student category, and some change both.

How should a confidence interval be read?

General explanation: a confidence interval describes a range of estimates compatible with the observed data under the statistical method used. It communicates uncertainty around an estimated difference, rather than giving a guarantee that the true difference lies inside the interval. An interval that crosses zero includes no difference as a compatible value; an interval entirely on one side of zero does not include zero. The interval's width is also informative: narrower intervals indicate greater precision, while wider intervals indicate less precision. The registry record does not report the method used to construct the posted intervals.

What does a p-value tell the reader?

General explanation: a p-value measures how unusual data at least as incompatible with a specified null hypothesis would be if that hypothesis and the analysis assumptions were correct. It is not the probability that the null hypothesis is true, and it does not measure the size or importance of a difference. The posted p-values are unadjusted. The registry record does not report whether the comparisons were prespecified, what inferential method was planned, or whether any adjustment for multiple comparisons was intended.

The comparison estimates include pooled standard deviations and standardized effect estimates. General explanation: a pooled standard deviation combines the within-group variation in the two groups being compared, and a standardized effect expresses a difference relative to that variability. Standardization can help describe the size of a contrast in relation to the spread of scores, but it does not remove confounding or turn a comparison of different student categories into a pure communication-format contrast. The registry record does not report an interpretation threshold for the standardized effects.

Why does group size matter?

General explanation: estimates based on smaller groups tend to be less precise because they contain less information about the range of individual outcomes. The two third-year medical groups have 20 and 19 participants in the primary outcome, while other groups have larger reported counts. This is reflected in the confidence intervals, which express uncertainty around each comparison. Group size alone does not determine whether an estimate is biased, and it does not establish whether the study was adequately planned.

For the primary outcome, the registry reports a Cohen's kappa of 0.74 for agreement between raters and describes that agreement as substantial. General explanation: kappa summarizes agreement beyond what would be expected from chance agreement under its assumptions. It concerns consistency of rating, not the size of the communication effect or the validity of the outcome for clinical care. The record says that one rater scored all protocols, a second scored a subset, and disagreements were resolved by consensus. It does not report the number of protocols independently rated or further details of the scoring rules.

Planning illustrations use the observed effect as a starting point and display scenarios targeting 100%, 75%, or 50% of that effect. General explanation: planning for a smaller effect requires more participants when other assumptions are held fixed, because a smaller signal is harder to distinguish from random variation. These calculations use the assumptions described with the planning estimates, rather than the trial's original design rationale. The original sample-size justification and original analysis methods are not reported in the registry record. No interim analyses or stopping rules are reported in the registry record.

10. Confidence Intervals and Precision

The confidence intervals show that precision differs across the posted comparisons. For the comparison of first-year psychology students in the non-structured condition with the structured reference group, the estimated mean difference is -2.41 items and the 95% confidence interval ranges from -4.02 to -0.800. Both ends are below zero, though the interval still spans a range of possible difference sizes.

For first-year medical students in the structured condition, the estimated difference from the reference is 1.83 items. Its interval is 0.127 to 3.533, with both ends above zero. The first-year medical non-structured comparison is estimated at 0.64 items, with an interval from -1.174 to 2.454. Because that interval crosses zero, it is compatible with a modest difference in either direction as well as no difference.

The third-year medical comparisons have estimates of 5.33 and 4.88 items, with intervals of 3.07 to 7.59 and 2.572 to 7.188, respectively. Both intervals remain above zero relative to the reference group. These estimates are based on 20 and 19 participants in the respective third-year groups, so their interpretation should keep the group sizes in view. The interval widths communicate uncertainty; they do not resolve whether student year, communication format, or both account for the observed contrasts.

General explanation: confidence intervals are more informative than a p-value alone because they display the range and direction of estimates compatible with the data under the method used. Precision is not the same as absence of bias, and the intervals do not account for unreported issues such as missing outcomes or choices among multiple analyses. The interval-construction method and any adjustments are not reported in the registry record.

11. Planning a Similar Trial: Sample Size

This is a planning illustration, not a re-evaluation of the original trial. It asks: if a new study wanted to detect a difference like the one observed, or a smaller one, how large would it need to be? It is deliberately not “observed power” of the finished trial, which is not a meaningful quantity.

The planning values are participants per arm, not total event counts. The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each scenario uses a stated fraction of the observed effect as the effect to detect. For time-to-event outcomes, planning values would represent total events, but the primary outcome in this study is not time-to-event, so its values are participants per arm.

The 80% power column describes the number of participants per arm estimated to give an 80% chance of detecting the chosen effect under the planning assumptions, if that effect and the assumptions are correct. The 90% power column describes the corresponding number per arm for a higher chance of detecting that same effect. These percentages are planning targets, not descriptions of how likely the observed result is to be true. This is not observed power.

Read each effect scenario as a planning illustration tied to the observed comparison: 100% uses the full observed effect, 75% uses a smaller fraction, and 50% uses a still smaller fraction. Smaller effects need more participants because they are harder to distinguish from random variation. The participant counts can therefore rise substantially in the smaller-effect scenarios, especially when the observed contrast is modest. The displayed estimates should not be treated as the trial's actual enrollment target or as a retrospective evaluation of its design. The registry record does not report the trial's original sample-size assumptions or justification, and these planning calculations do not fill that gap.

12. How the Sample Size Is Derived

Power is the probability that a study of a given size rejects the null hypothesis when a specific true effect exists. It depends on four things: the effect to detect, the variability of the outcome, the significance level (here a two-sided α of 0.05, so z_1−α/2 = 1.960) and the desired power (z_1−β = 0.842 for 80% and 1.282 for 90%).

Formula

n is the number of participants in each arm, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -2.410; pooled standard deviation σ = 4.098.
2. n = 2 × (1.96 + 0.842)² × 4.098² / -2.410² = 45.4, rounded up to 46 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 <- -2.41                      # observed difference in means
sigma <- 4.09750635549639                     # 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(1540266)
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 = 46 (80% power), 61 (90% power)
 75% of observed effect: n per arm = 81 (80% power), 108 (90% power)
 50% of observed effect: n per arm = 182 (80% power), 243 (90% power)
power.t.test: 46.4 
Simulated power at n = 46 per arm: 0.801 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-test rejects the null. It should land close to the target power; at very small required sample sizes it can fall below it, because the normal approximation is rough there.

13. Multiplicity, Missing Data and Analysis Populations

The posted results include five comparisons against the first-year psychology structured reference group, and the registry lists two secondary outcomes in addition to the primary outcome. It also reports three outcome measures. Multiple comparisons can increase the chance that at least one result appears notable by chance when many hypotheses are tested. General explanation: multiplicity procedures control or otherwise address that issue by adjusting inference or defining a testing strategy. Whether the study used any such procedure is not reported in the registry record. The posted p-values are unadjusted and should be interpreted as individual comparison results, not as a family-wise conclusion across outcomes and groups.

The registry record does not report a complete statistical analysis plan, a prespecified testing hierarchy, or which comparisons were primary, secondary, or exploratory beyond identifying the primary endpoint. It does not report whether analyses adjusted for student category, baseline characteristics, or other covariates. The reported contrasts are simple comparisons with a single reference group; no pooled structured-versus-non-structured estimate across student categories is posted.

Missing outcome data, exclusions after enrollment, and the analysis population are not reported in the registry record. The primary outcome counts differ by group, but the record does not explain whether that reflects missing scores, exclusions, group assignment, or another reason. There is no reported description of an intention-to-treat population, per-protocol population, or method for handling missing outcomes. The record also does not report imputation, sensitivity analyses, or reasons for withdrawals.

These omissions matter because the analysis population and missing-data strategy can affect the interpretation of group comparisons. The reported group means and estimates remain descriptive of the posted outcome data, but the registry record does not give enough information to assess how complete those data were or whether analysis choices were made before results were known.

14. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The registry reports zero serious adverse events and zero other adverse events among the participants listed as at risk in each of the six groups. The at-risk counts are 63 and 42 for the two first-year psychology groups, 60 and 38 for the two first-year medical groups, and 20 and 19 for the two third-year medical groups. The number of deaths is not reported in the registry record.

These entries describe adverse-event reporting for a study in which students watched communication videos. Zero recorded events in the posted categories means the registry entries list none as affecting the at-risk participants; it does not establish that every possible event was actively assessed or that no unreported event occurred. The registry record does not provide further details about event definitions, collection methods, follow-up duration for safety, or whether the study was designed to evaluate safety. The event counts should therefore be read as registry reporting, not as a general safety assessment of discharge communication or clinical treatment.

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

The posted results show variation in immediate recall across the six student groups. The first-year psychology group shown the structured video had a mean of 8.12 recalled items. Its non-structured counterpart had a lower mean of 5.71, and the unadjusted comparison interval lies below zero. The first-year medical structured group mean was 9.95, while its non-structured group mean was 8.76. The third-year medical group means were 13.45 in the structured condition and 13.0 in the non-structured condition. Both third-year groups had higher posted means than the first-year psychology structured reference group.

These figures describe observed performance in a controlled video task. The reference-group comparisons do not by themselves establish that structuring caused a change in recall across all students, because the contrasts also differ by field or year of study. The record does not report a direct pooled comparison of the two communication formats within or across all student groups, nor an adjusted analysis that separates these factors.

The outcome measures immediate recall at 5 minutes, not long-term retention, comprehension, adherence, patient experience, or clinical outcomes. The older patient in the video was played by an actor, and participants were students rather than patients receiving actual discharge communication. The results therefore do not directly describe what would happen in a real discharge setting.

The registry record does not report a minimally important difference, so the practical importance of any numerical difference cannot be determined from these results alone. The confidence intervals communicate statistical uncertainty around the posted comparisons, but they do not address unreported multiplicity, missing data, or analysis choices. The findings should be understood as group-level recall results from this study, with those limits in mind.

16. Why This Trial Matters Statistically

This study illustrates why clinical communication research can involve more than a simple two-group contrast. The reported groups combine video format with student discipline and year of study. Comparing each group with one reference group produces interpretable mean differences, but those contrasts answer different questions. A structured-versus-non-structured comparison within a student category is conceptually distinct from a comparison between first-year psychology and third-year medical students.

The outcome also shows how measurement choices shape interpretation. Recall was scored from protocols by independent raters, with disagreements resolved by discussion. The reported kappa of 0.74 gives readers information about agreement between raters, while the group means and standard deviations describe recall performance. These are separate properties: a rating process can show agreement, while group estimates still have uncertainty and may have limited relevance beyond the study task.

Statistically, the results provide a useful setting for examining means, standard deviations, confidence intervals, unadjusted p-values, and planning estimates. The first-year medical non-structured comparison has an interval crossing zero, while the other posted comparison intervals lie on one side of zero. That distinction is about uncertainty under each comparison, not proof of a global pattern or a clinical effect. The multiple groups and listed secondary outcomes also make the absence of a reported multiplicity strategy important when interpreting unadjusted tests.

Finally, the study separates a plausible mechanism—organization of information—from the more complex conditions of clinical care. Its recall task can describe immediate memory after a video, but it cannot by itself establish how communication affects patients in practice. Recognizing the scope of the outcome and the structure of the comparisons is central to drawing appropriately limited statistical conclusions.

17. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The principal limitation is the setting: participants watched a video rather than receiving discharge communication in a clinical encounter. The patient was an actor, and the participants were students. The measured endpoint was immediate recall at 5 minutes, so the results do not establish longer-term retention or patient-centered outcomes. The registry record does not report a minimally important difference for the recall measure.

The comparison structure also limits interpretation. The posted estimates compare each listed group with first-year psychology students in the structured condition. As a result, several contrasts vary in student category as well as communication condition. The registry record does not report a pooled comparison of structured versus non-structured communication or an adjusted model separating these factors.

Group sizes for the primary outcome vary, with 20 and 19 participants in the third-year medical groups. The reason for differences between the outcome counts and the reported adverse-event at-risk counts is not reported in the registry record. Missing outcome data, exclusions, withdrawals, and the analysis population are also not reported. These omissions make it difficult to assess completeness and the handling of participants without outcome scores.

The registry record identifies single masking but does not say who was masked. It does not report the randomization method, allocation concealment, the scoring rubric, the full analysis plan, or any multiplicity adjustment. The posted p-values are unadjusted. The detailed results for the two secondary outcomes are not reported in the registry record. These limitations do not erase the reported group summaries, but they constrain causal, clinical, and confirmatory interpretations.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Basel Discharge Communication Project — NCT01540266.
• Registry record last updated 2015-03-31; status last verified 2015-03.
• Results first posted 2015-03-31.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

21. Record Summary

NCT01540266 is the completed Basel Discharge Communication Project, Study 3, sponsored by University Hospital, Basel, Switzerland. The registry describes a randomized, single-masked, parallel-group study with 242 enrolled participants. First-year psychology students, first-year medical students and third-year medical students watched a physician present identical discharge information in structured or non-structured form.

The primary endpoint was the number of items recalled from the video at 5 minutes, reported as a mean and standard deviation. The video contained 28 information items. The primary outcome description reports Cohen's kappa of 0.74 for interrater agreement and describes consensus resolution of rating disagreements. Two secondary outcomes are listed, but their results are not reported in the registry record.

Posted comparisons use first-year psychology students in the structured condition as the reference and report unadjusted mean differences, confidence intervals, and p-values. A multiplicity strategy, missing-data approach, analysis population, and original sample-size justification are not reported in the registry record. The findings describe immediate recall in a student video task and do not directly establish outcomes in clinical discharge encounters.