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 The ISLAND Study: InSuLa Assessed Needs for Depression (NCT02137369), a randomized phase 4 study. The primary outcome is Number of Remission From Major Depressive Episode Events.
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT02137369 |
| Phase | Phase 4 |
| Condition | Depression |
| Allocation | Randomized |
| Design model | Crossover |
| Masking | Single |
| Enrollment | 77 |
| Arms in the primary outcome | 2 |
| Lead sponsor | Emory University |
| Results posted | Yes |
2. Clinical Question
Population
Male and female outpatients aged 21-55 years with major depression were eligible according to the study description. The registry lists depression as the condition and reports enrollment of 77 participants. The description anticipated enrolling up to 100 people.
Intervention
The initial treatment options were an SSRI or cognitive behavioral therapy (CBT). The SSRI could be escitalopram or sertraline. Nonresponders were described as crossing over to the alternative intervention for additional treatment. A combination-treatment intervention is also listed in the registry.
Comparator
The primary outcome groups are labelled SSRI and CBT. The registry describes randomized assignment to those initial treatments, but the posted primary outcome groups do not explain how crossover participants were handled in the outcome count.
Primary question
At 12 weeks, how many remission events from a major depressive episode were recorded in each treatment group, with remission assessed using the 17-item Hamilton Depression Rating Scale? The registry posts 15 events for SSRI and 4 for CBT, but does not report an inferential comparison of those counts.
The study’s broader aim was to test whether brain imaging findings might help guide treatment selection. The description refers to resting-state PET and BOLD fMRI scans before treatment and at the end of treatment, and to examining changes in brain function in participants receiving CBT or an SSRI. The primary result available in the registry record is a remission-event count, not a reported test of whether imaging predicts differential response or identifies which treatment is more suitable for an individual. The record does not report an imaging biomarker analysis or an estimate of its predictive performance.
3. Trial Design
The ISLAND Study is a phase 4 randomized trial with a crossover model and single masking, according to the registry. The record does not identify which participants or study personnel were masked. It lists Emory University as sponsor and gives a status of completed. The study start is recorded as 2014-09, and primary completion as 2019-10-24.
The study description says participants were randomized to escitalopram or CBT for 12 weeks. Participants who did not respond were to receive an additional 12 weeks of the alternative intervention. Participation was expected to last 14-26 weeks. The SSRI intervention description lists escitalopram, pill form, 20mg-40mg daily for 12 weeks, or sertraline, pill form, 50 - 150 mg daily for 12 weeks. CBT is described as 16 1-hour sessions over 12 weeks. The registry also lists combination treatment (SSRI + CBT), but the posted primary outcome groups are SSRI and CBT; the record does not provide a primary outcome result for a combination-treatment group.
Enrollment is recorded as 77, while the primary outcome results list 49 participants in the SSRI group and 11 in the CBT group. The registry does not explain the difference between the enrollment figure and the outcome-group counts, nor does it explain whether the group counts reflect eligibility, follow-up, treatment received, or another analysis population. It also does not describe how participants who crossed over were included in the primary outcome groups. These gaps matter when interpreting a randomized crossover study, because the intended assignment, actual treatment exposure, and outcome analysis population can differ. The trial protocol’s detailed allocation procedures and analysis plan are not reported in the registry record.
4. Participant Flow and Accounting
The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.
Before assignment: 77 were consented, 60 started treatments
Overall Study
| Stage | SSRI | Cognitive Behavioral Therapy |
|---|---|---|
| STARTED | 49 | 11 |
| COMPLETED | 41 | 9 |
| NOT COMPLETED | 8 | 2 |
Reasons for not completing this period:
| Reason | SSRI | Cognitive Behavioral Therapy |
|---|---|---|
| Withdrawal by Subject | 3 | 2 |
| Adverse Event | 2 | 0 |
| Lost to Follow-up | 1 | 0 |
| Moved away | 2 | 0 |
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 | SSRI | Cognitive Behavioral Therapy | Total |
|---|---|---|---|---|
| Participants analysed | 49 | 11 | 60 | |
| Age, Categorical: <=18 years | Participants | 0 | 0 | 0 |
| Age, Categorical: Between 18 and 65 years | Participants | 49 | 11 | 60 |
| Age, Categorical: >=65 years | Participants | 0 | 0 | 0 |
| Sex: Female, Male: Female | Participants | 35 | 8 | 43 |
| Sex: Female, Male: Male | Participants | 14 | 3 | 17 |
| Race (NIH/OMB): American Indian or Alaska Native | Participants | 2 | 0 | 2 |
| Race (NIH/OMB): Asian | Participants | 1 | 0 | 1 |
| Race (NIH/OMB): Native Hawaiian or Other Pacific Islander | Participants | 0 | 0 | 0 |
| Race (NIH/OMB): Black or African American | Participants | 11 | 4 | 15 |
| Race (NIH/OMB): White | Participants | 28 | 6 | 34 |
| Race (NIH/OMB): More than one race | Participants | 4 | 1 | 5 |
| Race (NIH/OMB): Unknown or Not Reported | Participants | 3 | 0 | 3 |
| Region of Enrollment: United States | participants | 49 | 11 | 60 |
6. Primary Endpoint
The registry lists the primary outcome as Number of Remission From Major Depressive Episode Events, measured over 12 weeks.
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | not applicable |
| Unit | number of events |
| Endpoint classification | not reduced to a single comparison |
The primary endpoint is the number of remission-from-major-depressive-episode events at 12 weeks. Remission was assessed using the 17-item Hamilton Depression Rating Scale. The posted result is 15 events in the SSRI group, with 49 participants listed, and 4 events in the CBT group, with 11 participants listed.
A remission endpoint addresses whether participants reached a defined low-symptom state, rather than measuring only average symptom change. For clinical research, that distinction can make the outcome easier to relate to a treatment goal, while still depending on the assessment method and the threshold used to define remission. The registry names the scale but does not report the remission threshold, assessment procedures, or whether the event count represents unique participants, repeated events, or another event-counting convention. Those details are not reported in the registry record.
The endpoint is posted as a count, not as an adjusted treatment effect or a rate with a confidence interval. The two group counts and their listed group sizes should therefore be read as descriptive results. They do not by themselves show the precision of a treatment contrast, account for crossover, or establish whether any observed difference could be explained by chance or other features of the analysis.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The posted primary outcome consists of 15 remission events for SSRI and 4 for CBT at 12 weeks. The corresponding listed group sizes are 49 and 11. These are descriptive counts. The registry does not post a treatment-effect estimate, confidence interval, or p-value for the primary outcome. It also records no analyses. A formal comparison of the two counts therefore cannot be interpreted from the posted results.
The counts should not be treated as directly comparable participant-level remission proportions without knowing the event-counting rules and analysis population. The endpoint is explicitly described as a number of events, and the registry does not say whether each event represents a different participant or how repeated events, if any, were handled. It also does not explain the difference between the outcome-group sizes or how crossover participants contributed to the counts. Without these details, readers cannot determine whether the groups represent original randomized assignments, treatment received, participants with complete assessments, or another grouping.
The design makes those distinctions important. Participants were described as randomized to an initial treatment, with nonresponders crossing over to the alternative intervention. A result based on original assignment would answer a different question from one grouped by treatment received or restricted to participants who completed a treatment period. The registry record does not specify which approach was used for the posted counts. It also does not report an analysis that accounts for the crossover structure.
The primary result is not an imaging result. The study description emphasizes whether brain scans might help select treatment, but no reported analysis links scan findings to remission or compares imaging-based treatment selection with another approach. The posted primary endpoint can describe remission events by listed treatment group, but it cannot establish the predictive value of an imaging biomarker.
The registry lists one secondary outcome, number of response-to-treatment events, but does not provide a secondary outcome result in the registry record for interpretation. It reports no analysis details for the primary or secondary outcomes. Thus, the record supports a limited descriptive reading of the remission counts and does not support claims about statistical significance, effect precision, subgroup differences, or predictive performance. Any treatment conclusion would go beyond what these posted statistics establish.
9. Statistical Methods Explained
The primary endpoint is a count of remission events. Count outcomes record how often an event occurs in a defined group or observation period. Their interpretation depends on what qualifies as an event, whether the same participant can contribute more than one event, and how follow-up time and missing assessments are handled. The registry gives the endpoint name, 12-week timeframe, and assessment scale, but does not report the event-counting rules or a statistical model for the count.
What can a count tell readers?
A count is a direct descriptive summary, but it does not automatically provide a fair comparison between groups of different sizes. A treatment comparison generally needs a clearly defined estimand and an analysis that accounts for the number of participants at risk or the time over which events could occur. For a count of remission events, a model might also need to address whether events are independent and whether the outcome reflects one event per participant. These are general statistical considerations, not descriptions of analyses reported by this trial. The registry does not identify a count model, an offset, an adjustment set, or a prespecified contrast.
Over-dispersion is a general issue for count data. It occurs when the observed variability is greater than a chosen model expects. If that feature is present, a model that assumes less variability can produce confidence intervals that are too narrow and tests that are too optimistic. The registry does not report whether over-dispersion was assessed, whether a model was fitted, or how uncertainty was estimated. It is therefore not possible to infer that any particular count-data method was used.
Why does the crossover design matter?
In a crossover design, participants may receive more than one treatment over time. The description specifies an initial randomized treatment and crossover for nonresponders, but the posted primary endpoint is framed as a 12-week count. The record does not say whether the result is based only on the initial treatment period, includes later crossover treatment, or classifies participants according to treatment received. These choices affect what contrast the analysis represents. They can also affect whether observations are independent, because outcomes from the same participant are related. The registry does not report a method for handling within-participant dependence or period effects.
Randomization is intended to make assigned groups comparable at the start of a trial, on average. That advantage is most directly retained by an analysis based on randomized assignment, with a clearly defined population and appropriate handling of outcome availability. The record does not state whether the remission counts follow that approach. It gives different group counts for the primary outcome and does not explain how those analysis groups were formed. No conclusion about adherence to an intention-to-treat or other analysis population can be drawn.
What does a confidence interval add?
In general, a confidence interval describes the uncertainty around an estimated effect under the assumptions of the analysis. It can show both the range of effects reasonably compatible with the data and whether the estimate is precise or imprecise. A count in each group alone does not provide that information. The registry record does not report a confidence interval or a model-based estimate for the primary outcome. It also does not report a p-value. The absence of those statistics means that the posted counts cannot be used to assess statistical compatibility with a null effect through a reported inferential procedure.
The record includes no analysis count, and the stated reason for omitting a two-arm comparison is that values were missing or not estimable for a count-other endpoint. Consequently, there is no sponsor-reported inferential result to interpret. The simple recalculations ordinarily used for unadjusted comparisons are not available as a valid treatment contrast for this endpoint in the posted results. A calculation from incomplete or ambiguous summaries would not resolve the missing definition of the event count or the crossover analysis population.
How should planning illustrations be understood?
Planning sample sizes are not observed trial results and are not observed power. The registry’s stated planning assumptions refer to a two-sided significance level of 0.05, equal allocation, a normal approximation, and detecting a stated fraction of the observed effect. Those assumptions describe a planning illustration, not the trial’s own design justification. The registry does not report the protocol’s sample-size rationale or a separate justification for the achieved enrollment. No planning sample-size values are included in the record details interpreted. In general, a planning calculation depends on its assumptions and should not be read as evidence that the study had a particular probability of finding an effect after the data were collected.
These limitations are not a judgment about the study’s conduct. They delimit what the registry results can support: descriptive event counts are posted, while the endpoint definition, comparison method, uncertainty estimate, and handling of the crossover structure are not reported in enough detail to reconstruct an inferential analysis.
10. Multiplicity, Missing Data and Analysis Populations
The registry lists a primary outcome of remission events at 12 weeks and one secondary outcome, response-to-treatment events. It reports two outcome measures overall, but no analyses. No multiplicity procedure, hierarchy among outcomes, or adjustment for multiple testing is reported in the registry record. Without a stated testing plan and inferential results, it is not possible to determine whether any reported analysis controlled the chance of false-positive findings across outcomes or other comparisons.
Missing-data methods are also not reported. The record does not state how missed assessments, withdrawals, incomplete treatment periods, or unavailable scans were handled. It does not describe imputation, sensitivity analyses, or rules for classifying participants without a 12-week assessment. The primary outcome group counts are 49 for SSRI and 11 for CBT, while enrollment is 77; the registry does not explain the relationship among these figures. That gap prevents a clear assessment of outcome availability or attrition by assigned group.
The analysis population is not described. The registry does not say whether the remission counts use all randomized participants, participants who received treatment, participants who completed the initial treatment period, or another subset. It also does not explain how participants who crossed over after nonresponse were classified. These omissions limit interpretation of the group counts and make it difficult to judge whether the analysis preserves the original randomized comparison. The protocol’s missing-data plan, analysis-population definitions, and multiplicity strategy are 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 |
|---|---|---|---|
| SSRI | 1 | 49 | 2.0% |
| Cognitive Behavioral Therapy | 0 | 11 | 0.0% |
Deaths
| Arm | Deaths (participants affected of at risk) |
|---|---|
| SSRI | 0 of 49 |
| Cognitive Behavioral Therapy | 0 of 11 |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| SSRI | 44 of 49 |
| Cognitive Behavioral Therapy | 5 of 11 |
Most frequent serious adverse event terms
| Event term | Organ system | SSRI | Cognitive Behavioral Therapy |
|---|---|---|---|
| Hospitalization for suicidal ideation | Psychiatric disorders | 1 of 49 | 0 of 11 |
Most frequent non-serious adverse event terms
| Event term | Organ system | SSRI | Cognitive Behavioral Therapy |
|---|---|---|---|
| Headache | General disorders | 19 of 49 | 1 of 11 |
| Nausea | Gastrointestinal disorders | 12 of 49 | 0 of 11 |
| Upper respiratory tract infection | Respiratory, thoracic and mediastinal disorders | 8 of 49 | 2 of 11 |
| Insomnia | Nervous system disorders | 9 of 49 | 0 of 11 |
| Anxiety | Psychiatric disorders | 9 of 49 | 0 of 11 |
| Musculoskeletal pain | Musculoskeletal and connective tissue disorders | 8 of 49 | 1 of 11 |
| Sedation | Surgical and medical procedures | 8 of 49 | 1 of 11 |
| Diarea | Gastrointestinal disorders | 8 of 49 | 0 of 11 |
| Reduced Concentration | General disorders | 7 of 49 | 0 of 11 |
| Fatigue | General disorders | 7 of 49 | 0 of 11 |
Serious events with the highest arm-level rate: hospitalization for suicidal ideation (2.0%).
The registry reports serious adverse events among 49 participants at risk in the SSRI group and 11 in the CBT group. One participant in the SSRI group and no participants in the CBT group are listed as affected by a serious adverse event. Deaths are listed as 0 in each group. The most frequently listed serious-event term is hospitalization for suicidal ideation, with 2.0% recorded. The registry does not assign that term to a specific treatment group in the details.
These figures describe the adverse-event entries in the registry; they do not establish that an event was caused by treatment. The record does not provide event-level descriptions, timing, adjudication procedures, or a causal assessment. The number of participants at risk is not the same as a complete account of exposure or follow-up time, and the registry does not explain the safety analysis population or how crossover was handled. The listed group counts should therefore be read as recorded safety summaries, not as a comparative estimate of treatment-related risk.
The record also lists other adverse events affecting 44 participants in the SSRI group and 5 in the CBT group. These are distinct from the serious adverse-event figures. They should not be combined with the serious-event counts or treated as evidence of severity without further detail. No advice about treatment choices follows from these registry summaries.
12. What the Result Does, and Does Not, Mean
The registry record documents a randomized, single-masked, phase 4 study with a crossover design and reports primary outcome counts for remission at 12 weeks. Those counts are 15 in the SSRI group and 4 in the CBT group. The study description identifies remission as assessed by the 17-item Hamilton Depression Rating Scale. These registry entries provide a descriptive account of the posted outcome, but the record does not provide an inferential comparison or the details needed to reconstruct one.
The results do not establish that one treatment produced more remission than the other. The counts are not accompanied by an estimated contrast, uncertainty interval, or p-value, and their group sizes differ. The record does not explain how the groups were defined, how crossover participants were handled, or whether each event represents a unique participant. Without that information, the counts do not define a clear treatment-effect estimate.
The results also do not answer the study’s broader imaging question. The description says the study aimed to test whether brain scan findings might help physicians select treatment, and it describes PET and fMRI assessments. However, no biomarker analysis or treatment-selection result is posted in the registry record for interpretation. The remission counts alone cannot establish whether imaging predicts response or improves treatment stratification.
Nor do the posted figures show the precision of an effect or settle whether a treatment difference is clinically important. Those questions require a well-defined comparison, analysis population, and uncertainty estimate. Missing information about outcome ascertainment, missing data, and the crossover analysis further limits interpretation. The appropriate conclusion is narrow: remission-event counts are recorded for the two outcome groups, but the registry does not report a statistical analysis that supports a comparative or predictive claim.
13. Why This Trial Matters Statistically
This trial raises several important statistical issues because its scientific aim and its posted primary result are not the same type of question. The study description focuses on whether imaging might help stratify treatment, while the primary outcome is a count of remission events. A descriptive outcome count can document what was recorded, but a biomarker-guided treatment question requires an analysis of how imaging findings relate to outcomes across treatment options. No such result is reported in the registry record.
The randomized crossover structure also makes analysis choices consequential. Participants were assigned to an initial treatment, and nonresponders were expected to cross over. The primary outcome is specified at 12 weeks, but the record does not clarify the treatment period or participant grouping used for the posted counts. A clear estimand—what treatment effect is intended to be estimated, for which participants, and over what period—is needed to interpret a treatment comparison. Without it, even a numerical contrast could be difficult to connect to the study question.
Finally, the mismatch between the overall enrollment and the listed primary outcome group sizes, together with absent analysis details, illustrates why results need more than counts. Definitions of the analysis population, event-counting rules, missing-data handling, and uncertainty estimates are essential for judging the strength and limits of evidence. The registry record provides a useful description of the trial and its posted counts, while leaving these statistical details unresolved.
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 primary outcome is reported as an event count, but the registry does not define whether each event corresponds to a distinct participant or whether repeat events are possible. It names the 17-item Hamilton Depression Rating Scale but does not state the remission threshold or the assessment procedures. These details are needed to understand exactly what the endpoint represents.
The record does not explain why the primary outcome lists 49 participants for SSRI and 11 for CBT when enrollment is 77. It also does not clarify whether those groups reflect randomized assignment, treatment received, completion of the initial treatment period, or availability of outcome data. Because nonresponders were expected to cross over, the absence of an explanation for their classification is particularly relevant to interpreting a 12-week result.
No primary-outcome analysis, effect estimate, confidence interval, or p-value is reported. The registry’s stated reason for not presenting a two-arm comparison is that values were missing or not estimable for a count-other endpoint. No statistical model, covariate adjustment, missing-data method, or analysis-population definition is described. The registry also does not report an imaging-based treatment-selection result, despite the study’s stated imaging aim.
Additional protocol details, including the sample-size justification, are not reported in the registry record. Masking is identified as single, but the masked party is not specified. The combination-treatment intervention is listed without a corresponding primary outcome group result. These limitations do not determine how the study was conducted; they identify information the registry record does not report and that is necessary for a fuller statistical interpretation.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: The ISLAND Study: InSuLa Assessed Needs for Depression — NCT02137369.
- Registry record last updated 2020-11-05; status last verified 2020-10.
- Results first posted 2020-11-05.
- 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
The ISLAND Study, NCT02137369, is a completed phase 4 randomized trial sponsored by Emory University. It is listed as single-masked and crossover, with depression as the condition and enrollment of 77. The study description says participants were initially randomized to an SSRI or CBT, with nonresponders crossing over to the alternative intervention. The interventions also list combination treatment.
The primary outcome is the number of remission events from a major depressive episode at 12 weeks, assessed using the 17-item Hamilton Depression Rating Scale. The posted counts are 15 for SSRI, with 49 participants listed, and 4 for CBT, with 11 participants listed. No inferential analysis is reported. The registry does not explain the outcome-group sizes, event-counting rules, crossover classification, or missing-data handling. It also does not report an imaging biomarker analysis, although treatment stratification using imaging is the study’s stated broader aim.