This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Goals of Care: A Nursing Home Trial of Decision Support for Advanced Dementia (NCT01565642), a randomized na study. The primary outcome is Quality of Communication and Decision-making.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT01565642 |
| Phase | NA |
| Condition | Dementia |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | Triple |
| Enrollment | 302 |
| Arms in the primary outcome | n/a |
| Lead sponsor | University of North Carolina, Chapel Hill |
| Results posted | Yes |
2. Clinical Question
Population
Nursing home residents with advanced dementia are the population identified in the trial summary. Dementia is the listed condition. The registry record does not give further eligibility details or describe the population included in the primary outcome analysis.
Intervention
The intervention is goals-of-care decision support, described as a decision aid and care plan meeting. The trial summary frames the intervention as support for communication and decision-making about goals of care and palliative care.
Comparator
The control condition is attention control information on dementia care. The record does not give further detail about the content or delivery of this information.
Primary question
The primary question concerns whether decision support affects quality of communication and decision-making, assessed using the Quality of Communication score and its End of Life subscale at 3 months. Higher scores indicate better communication quality. Primary outcome results are not reported in the registry record, so the direction and precision of any difference cannot be described.
3. Trial Design
The registry record describes a completed, randomized, parallel, triple-masked, cluster-randomized controlled trial sponsored by the University of North Carolina, Chapel Hill. The intervention is behavioral: goals-of-care decision support delivered as a decision aid and care plan meeting. The comparator is attention control information on dementia care. The record gives enrollment as 302 and lists dementia as the condition.
The listed start date is 2012-05, and primary completion is listed as 2015-07. The study status is completed. Phase is recorded as NA. These administrative details describe the trial record, but they do not establish when individual outcomes were measured beyond the primary outcome timeframe stated as 3 months.
Cluster randomization means that assignment occurs at a group level rather than by treating every participant as an independent randomization unit. The registry record does not identify the cluster unit or give details about the numbers or sizes of clusters. This matters statistically because outcomes within a cluster may be more alike than outcomes from unrelated clusters. An analysis that treats every person as independent can therefore misrepresent uncertainty. The record specifically indicates that a simple individual-level re-analysis would be misleading.
The masking label is triple, but the registry record does not identify which roles were masked or describe how masking was maintained. It also does not report the allocation procedure beyond stating that allocation was randomized. The record does not provide a protocol description, sample-size justification, or details of the statistical analysis plan. No primary outcome estimates or analysis records are reported, so design characteristics alone cannot establish the size or precision of an intervention effect.
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
| Stage | Decision Support Intervention | Control |
|---|---|---|
| STARTED | 151 | 151 |
| COMPLETED | 149 | 150 |
| NOT COMPLETED | 2 | 1 |
Reasons for not completing this period:
| Reason | Decision Support Intervention | Control |
|---|---|---|
| Withdrawal by Subject | 2 | 1 |
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.
| Characteristic | Unit | Decision Support Intervention | Control | Total |
|---|---|---|---|---|
| Participants analysed | 151 | 151 | 302 | |
| Age, Continuous | years | 85.7 (SD 7.6) | 87.3 (SD 6.7) | 86.5 (SD 6.9) |
| Sex: Female, Male: Female | Participants | 121 | 125 | 246 |
| Sex: Female, Male: Male | Participants | 30 | 26 | 56 |
| Race/Ethnicity, Customized: Race – African American | Participants | 14 | 25 | 39 |
| Race/Ethnicity, Customized: Race – White | Participants | 134 | 123 | 257 |
| Race/Ethnicity, Customized: Race – Other | Participants | 3 | 3 | 6 |
| Region of Enrollment: United States | Participants | 151 | 151 | 302 |
| Dementia Stage: Global Deterioration Scale (GDS): GDS 5 | Participants | 38 | 36 | 74 |
| Dementia Stage: Global Deterioration Scale (GDS): GDS 6 | Participants | 79 | 73 | 152 |
| Dementia Stage: Global Deterioration Scale (GDS): GDS 7 | Participants | 34 | 42 | 76 |
6. Primary Endpoint
The registry lists the primary outcome as Quality of Communication and Decision-making, measured over 3 months.
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | Standard Deviation |
| Unit | units on a scale |
| Endpoint classification | Continuous Endpoint |
The primary outcome is titled Quality of Communication and Decision-making. It uses the Quality of Communication (QOC) score and its End of Life (EOL) subscale score, reported in units on a scale. The registry describes scores ranging from 0-10 on the QOC and its subscales, with higher scores indicating better communication quality. The stated timeframe is 3 months.
In general, a scale outcome can represent a concept that is not directly observed as a single physical quantity. A questionnaire or scoring instrument translates responses into a numerical score, allowing outcomes to be described and compared across groups. A mean summarizes the average score, while a standard deviation describes how spread out individual scores are around that average. The registry lists the primary outcome parameter as a mean and its dispersion as a standard deviation.
Communication and decision-making are relevant to the stated trial aim because the intervention is designed to support goals-of-care discussions. The EOL subscale focuses the measurement on communication concerning end-of-life care, as indicated by its name and its inclusion in the primary outcome description. The registry record does not describe the instrument’s items, scoring details beyond the stated range and direction, who completed it, or how missing responses were handled.
The registry record does not report the observed group means, a between-group estimate, a confidence interval, or a p-value for this outcome. Consequently, the endpoint definition explains what the trial intended to assess, but it does not establish what the outcome data showed.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record gives a clear description of the intervention comparison and the intended primary outcome, but it does not report numerical primary outcome results. There are no group means for the QOC score or EOL subscale, no between-group estimate, and no confidence interval or p-value. The record also lists zero analyses. As a result, the size, direction, and precision of the primary outcome contrast cannot be interpreted from the registry results.
This absence is different from a result showing no difference. A result close to zero with a sufficiently narrow confidence interval could indicate that large differences are less compatible with the data. A missing estimate gives no such information. It does not demonstrate that the intervention and control groups had the same scores, nor does it establish that either group had better communication quality. The record’s description of the outcome scale gives the meaning of higher scores, but without observed scores that direction cannot be applied to the trial comparison.
The trial is cluster randomized. In general, people assigned within the same cluster can have correlated outcomes because they share settings or other cluster-level circumstances. Accounting for that dependence is necessary to obtain suitable uncertainty estimates. The registry record cautions that a simple individual-level re-analysis would be misleading. Accordingly, no unadjusted participant-level estimate should be treated as a substitute for an analysis that respects the cluster design. The registry record does not report the sponsor’s primary outcome estimate or the details needed to evaluate the design-aware analysis.
The listed secondary outcomes include palliative care domains in the care plan, satisfaction with care, comfort in dying, Alzheimer disease-related quality of life, quality of dying, frequency of communication, hospice referral, and hospitalizations. Their inclusion identifies other outcomes of interest; it does not supply results for them. Outcome counts and analysis counts do not replace outcome estimates or explain how endpoints were handled.
The record therefore supports a description of the research question, design, and outcome definitions, but not a conclusion about the observed effect. No effect size, statistical significance, or clinical importance can be established from the posted results. Even if estimates were available, statistical uncertainty and the relevance of a score difference would need to be considered separately.
9. Statistical Methods Explained
The registry identifies a randomized parallel design with cluster-level assignment. In general, randomization is intended to make comparison groups more comparable with respect to both measured and unmeasured factors, on average. It does not guarantee that groups will be identical, and the design alone does not show what happened to the outcome. The record does not describe the randomization sequence, concealment procedures, or the identities of the clusters.
Why does cluster randomization affect analysis?
When assignment is made by cluster, outcomes from people in the same cluster may be correlated. This is called within-cluster dependence. In general, positive dependence means that observations contain less independent information than the same number of unrelated observations would. An analysis that ignores this dependence can make standard errors too small and confidence intervals too narrow. The registry record explicitly warns that a simple individual-level re-analysis would be misleading. It does not report the cluster unit, cluster counts, cluster sizes, or the method used in the sponsor’s analysis, so the adequacy of any particular adjustment cannot be assessed from the record.
What would a mean and standard deviation describe?
The primary outcome is a continuous scale score, and the registry lists the mean as the outcome parameter and the standard deviation as its dispersion. In general, a mean describes the central average of observed scores, while a standard deviation describes how variable those scores are around the mean. They answer different questions: the mean concerns the average level, and the standard deviation concerns spread among observations. The record does not report either quantity for the intervention or control group.
A comparison of group means would ordinarily be accompanied by an estimated difference and an uncertainty interval. In general, a confidence interval gives a range of effect values compatible with the observed data and the assumptions of the analysis at its stated confidence level. Its width reflects precision: wider intervals indicate more uncertainty than narrower intervals. A confidence interval is not a guarantee that the true effect lies within its endpoints. No interval, estimate, or confidence level for the primary outcome is reported in the registry record.
What does a p-value address?
In general, 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 true. It does not measure the probability that the null hypothesis is true, the probability that the finding is due to chance, or the size or importance of an effect. A p-value is most informative when considered alongside an effect estimate, an uncertainty interval, the design, and the analysis plan. No primary outcome p-value is reported in this record, so statistical significance cannot be assessed.
Why does the analysis plan matter?
Trials may use different analysis choices, including adjustment for design features, covariates, or repeated measurements. Those choices can affect estimates and their uncertainty. For a cluster-randomized trial, the analysis needs to account for clustering. The registry record does not provide a statistical analysis plan, covariate list, missing-data method, or information about interim analyses. None of these details should be inferred from the title or from the fact that the study was randomized.
The record gives enrollment as 302, but that number alone does not describe how many observations contributed to the primary outcome analysis or how participants were distributed across clusters. It also does not establish whether the enrollment was sufficient to detect a particular effect. A sample-size justification is not reported in the registry record. Any planning illustration based on an observed effect is not observed power and cannot substitute for the trial’s design rationale or an outcome analysis.
What can be inferred from the registry entries?
The record’s zero analysis count and absence of primary outcome estimates mean that there is no posted inferential result to explain. This is a reporting limitation, not evidence for a null effect. The record identifies a primary measure and timeframe, but does not state the analyzed population, the handling of missing scores, the number of outcome observations, or whether the primary analysis followed the original assignment. The sponsor’s own analysis is the relevant basis for inference in a cluster-randomized trial; however, its numerical findings and methods are not reported in the registry record.
In general, analysis populations answer which participants or clusters contribute to an estimate. An intention-to-treat analysis, for example, is commonly defined around original randomized assignment, but the registry record does not say that such an analysis was used. It would be inappropriate to assign that label without documentation. Likewise, no adjustment or sensitivity analysis should be attributed to this trial without a record of it.
Overall, the statistical methods information allows readers to recognize the key design issue—cluster-level assignment—but does not permit a numerical re-analysis or an appraisal of the sponsor’s statistical model. The primary outcome result, its precision, the treatment of clustering, and the analysis population are all not reported in the registry record.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists a primary outcome and a set of secondary outcomes, including care-plan domains, satisfaction with care, comfort in dying, Alzheimer disease-related quality of life, quality of dying, communication frequency, hospice referral, and hospitalizations. Considering several outcomes can raise multiplicity concerns: in general, the more statistical comparisons made, the greater the chance that at least one may appear notable by chance if each is interpreted separately. The registry record does not report a multiplicity adjustment, a hierarchy among secondary outcomes, or a prespecified testing strategy. It is therefore not possible to determine how multiple comparisons were handled.
The record does not report primary or secondary outcome estimates, analysis populations, or outcome-specific participant counts. It does not describe missing-data frequency, reasons for missingness, or methods used to handle missing observations. These details matter because an analysis based only on observed scores can be biased if the availability of measurements is related to outcome or treatment assignment. In general, missing-data methods rely on assumptions, and those assumptions should be described rather than presumed.
The record also does not report whether analyses followed randomized assignment, whether exclusions occurred after randomization, or whether any sensitivity analyses were conducted. The masking label does not answer these analysis questions. Without a posted protocol or analysis plan, the intended handling of multiplicity, missing data, and analysis populations is not reported in the registry record.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Decision Support Intervention | 0 | 151 | 0.0% |
| Control | 0 | 151 | 0.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| Decision Support Intervention | 27 of 151 |
| Control | 33 of 151 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 1% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Decision Support Intervention | 0 of 151 |
| Control | 0 of 151 |
The adverse-event entries list zero serious adverse events for both the Decision Support Intervention and Control groups. The listed at-risk counts are 151 in each group. The record separately lists deaths as 27 in the Decision Support Intervention group and 33 in the Control group.
These are descriptive registry counts, not a comparative safety analysis. The record does not report statistical comparisons, confidence intervals, event timing, causes of death, or whether deaths were considered related to the intervention. It also does not explain the relationship between the serious-event and death categories or provide details about ascertainment and follow-up. The zero serious adverse-event counts should therefore be read as the entries recorded for that category, not as proof that no safety events of any kind occurred.
Death counts are not, by themselves, evidence that the intervention caused or prevented deaths. Interpreting them as a between-group effect would require information about outcome definitions, follow-up, exposure, and an analysis appropriate to the randomized cluster design. Those details and analyses are not reported in the registry record. The listed counts do not support a causal conclusion or a claim of equivalent safety.
12. What the Result Does, and Does Not, Mean
The trial addresses an important research question: whether goals-of-care decision support can improve communication and decision-making and the quality of palliative care for nursing home residents with advanced dementia. The intervention is described as a decision aid and care plan meeting, and the comparator as attention control information on dementia care. The primary outcome is defined as the QOC score and its EOL subscale at 3 months, with higher scores representing better communication quality.
What the registry record does not establish is the effect observed in the trial. It contains no primary outcome means, between-group estimate, confidence interval, or p-value. It also contains no analysis records. Thus, the posted results do not show whether scores differed, in which direction any difference went, or how precise an estimate might be. The absence of estimates must not be interpreted as evidence of no effect.
Even if an estimate were available, a statistical difference would not automatically establish that the difference was important to residents, families, or care teams. Statistical significance and practical or clinical relevance are distinct considerations. Interpreting relevance would require the observed magnitude and context, and the registry record does not report a threshold for an important score difference.
The cluster-randomized design is central to interpretation. General statistical reasoning requires accounting for similarity among people in the same cluster. The registry cautions against a simple individual-level re-analysis, and no sponsor analysis details are reported. Consequently, an improvised comparison of individuals would not provide a valid replacement for a design-aware analysis.
The record can support statements about the trial’s purpose, design labels, intervention and control descriptions, endpoint definition, listed secondary outcomes, and adverse-event counts. It cannot support a conclusion that the decision support improved, worsened, or left unchanged communication or palliative care outcomes. Those conclusions require outcome results and an appropriate analysis, neither of which is reported in the registry record.
13. Why This Trial Matters Statistically
This trial is statistically notable because it uses cluster randomization to evaluate a behavioral intervention in nursing homes. In cluster-randomized research, assignment at a group level can be appropriate to the way an intervention is organized, but it changes how outcomes should be analyzed. Within-cluster dependence can affect the amount of independent information and the precision of estimates. The registry itself warns that a simple individual-level re-analysis would be misleading.
The primary outcome is a scale measure of communication quality, including an EOL subscale. Such measures allow a complex concept to be represented quantitatively, but a score comparison needs more than a direction-of-scale statement. Readers also need group summaries, an effect estimate, uncertainty, and an analysis that respects the design. These primary results are not reported in the registry record.
The record also illustrates why design descriptions and outcome reporting serve different purposes. Randomization, masking, the intervention description, and the endpoint definition help explain how a study was intended to work. They do not reveal what the measured data showed. Enrollment and completion status likewise do not substitute for an analysis.
Finally, the trial’s broad outcome set makes transparent reporting of outcome selection and multiplicity relevant. The record identifies secondary outcomes but does not report a testing strategy, missing-data methods, or an analysis plan. These omissions limit assessment of how comprehensively and consistently the outcomes were evaluated. The statistical importance of the study therefore lies in both its cluster design and the need for complete, design-aware results to interpret its central question.
14. 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 for interpreting this record is that primary outcome results are not reported. There are no group means, treatment contrast, confidence interval, or p-value for the QOC score or EOL subscale. The record lists zero analyses, so no inferential finding can be assessed. This prevents evaluation of effect direction, magnitude, precision, or statistical significance.
Important design and analysis details are also absent. Although the trial is identified as cluster randomized, the registry record does not identify the cluster unit, cluster counts or sizes, or the approach used to account for clustering. The record indicates that a simple individual-level re-analysis would be misleading. Without the sponsor’s numerical analysis and methods, readers cannot evaluate whether the analysis reflected the design.
The record does not report the primary outcome analysis population, the number of observed scores, missing-data patterns, or missing-data methods. It does not provide a protocol or statistical analysis plan, sample-size justification, or details of interim analyses. It also does not explain who was masked despite the triple-masking label, or provide further eligibility and outcome-collection details.
Safety interpretation is limited as well. Serious adverse-event counts and death counts are listed, but event timing, causes, attribution, follow-up, and comparative analyses are not reported. The registry record does not provide enough context to treat those descriptive counts as evidence of a causal safety difference.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Goals of Care: A Nursing Home Trial of Decision Support for Advanced Dementia — NCT01565642.
- Registry record last updated 2018-08-10; status last verified 2016-03.
- Results first posted 2018-08-10.
- 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.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
NCT01565642 is a completed, randomized, parallel, triple-masked, cluster-randomized trial sponsored by the University of North Carolina, Chapel Hill. It evaluates goals-of-care decision support, described as a decision aid and care plan meeting, against attention control information on dementia care. The listed population is nursing home residents with advanced dementia, and enrollment is 302.
The primary outcome is the QOC score and its EOL subscale at 3 months. Scores range 0-10, with higher scores indicating better communication quality. The registry record lists secondary outcomes spanning care planning, satisfaction, comfort, quality of life and dying, communication, hospice referral, and hospitalizations.
Primary outcome estimates and analyses are not reported in the registry record. The record cautions that a simple individual-level re-analysis would be misleading because of cluster randomization. Serious adverse-event counts are listed as zero in both groups; deaths are listed as 27 and 33, respectively. These descriptive entries do not provide a comparative safety analysis.