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 A Duloxetine Dosing Strategy Study in Korean Patients With Major Depressive Disorder (NCT00960986), a randomized phase 4 study. The primary outcome is Mean Maximum Nausea Severity, Association for Methodology and Documentation in Psychiatry (AMDP-5) Adverse Event (AE) Scale Item 112 (Nausea).
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00960986 |
| Phase | Phase 4 |
| Condition | Major Depressive Disorder (MDD) |
| Allocation | Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 249 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Eli Lilly and Company |
| Results posted | Yes |
2. Clinical Question
Population
Korean patients with major depressive disorder. The primary outcome’s safety population included participants who took at least 1 study drug dose; they were analysed according to the drug dose actually taken. The registry record does not report further eligibility details in the information available for this analysis.
Intervention
The intervention was oral duloxetine hydrochloride, given once daily under one of four strategies. Two strategies used 60 mg throughout 8 weeks, with or without food. The other two began at 30 mg for 1 week, then used 60 mg for 7 weeks, with or without food.
Comparator
The four strategies can be considered in relation to one another: 30 mg followed by 60 mg with food; 60 mg with food; 30 mg followed by 60 mg without food; and 60 mg without food. These are comparisons between dosing strategies, not simply between fixed doses, because the 30 mg strategies changed to 60 mg after 1 week.
Primary question
The study asked how nausea severity differed across the four dosing strategies. Nausea was assessed with item 112 of the AMDP-5 adverse event scale, which ranges from 0, not present, to 3, severe. The primary measure was mean maximum nausea severity, with assessment timeframes at 1 week and 8 weeks. Numerical group results and between-strategy comparisons are not reported in the registry record.
3. Trial Design
This was a completed, randomized, open-label, parallel-group phase 4 study of duloxetine dosing strategies in Korean patients with major depressive disorder. The registry lists enrollment of 249. It records no masking, so participants and study staff were not blinded according to the design description. The randomization method, allocation concealment, and details of how participants were assigned among the groups are not reported in the registry record.
The four groups differed by dose strategy and food instruction. The 60 mg with-food and 60 mg without-food groups took 60 mg once daily for 8 weeks. The two 30 mg groups took 30 mg once daily for 1 week, then 60 mg once daily for 7 weeks, either with or without food. Thus, the study compared full regimens rather than maintaining a 30 mg dose for the entire study period.
The stated start was 2009-08, and primary completion was 2011-02. The registry status is completed. The record does not report the recruitment locations beyond describing the participants as Korean, nor does it provide additional detail about eligibility criteria, follow-up after treatment, or the conduct of randomization.
Because the study was open-label and the primary measure involved reported nausea severity, awareness of the assigned regimen could be relevant when interpreting subjective symptom ratings. This is a general design consideration, not evidence that ratings were biased in this study. Without numerical outcome results or a reported analysis plan, the registry record does not establish the size or precision of any difference between regimens.
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: Period 1 was a 3- to 30-day screening and washout period; Period 2 (Week 0-1) was a 1-week initial dosing period (randomization to duloxetine 30 mg with food, 30 mg without food, 60 mg with food, or 60 mg without food); Period 3 (Week 1-8) was a 7-week therapy period (treatment switched to duloxetine 60 mg once daily (QD) until study end.
Period 1 (Screening and Washout)
| Stage | Duloxetine 60 mg With Food | Duloxetine 60 mg Without Food | Duloxetine 30 mg With Food | Duloxetine 30 mg Without Food |
|---|---|---|---|---|
| STARTED | 59 | 63 | 63 | 64 |
| COMPLETED | 56 | 59 | 59 | 61 |
| NOT COMPLETED | 3 | 4 | 4 | 3 |
Reasons for not completing this period:
| Reason | Duloxetine 60 mg With Food | Duloxetine 60 mg Without Food | Duloxetine 30 mg With Food | Duloxetine 30 mg Without Food |
|---|---|---|---|---|
| Adverse Event | 1 | 2 | 1 | 1 |
| Lost to Follow-up | 0 | 0 | 2 | 1 |
| Protocol Violation | 1 | 0 | 0 | 0 |
| Withdrawal by Subject | 1 | 2 | 1 | 1 |
Period 2 (1-week Initial Dosing Period)
| Stage | Duloxetine 60 mg With Food | Duloxetine 60 mg Without Food | Duloxetine 30 mg With Food | Duloxetine 30 mg Without Food |
|---|---|---|---|---|
| STARTED | 56 | 59 | 59 | 61 |
| COMPLETED | 56 | 59 | 59 | 61 |
| NOT COMPLETED | 0 | 0 | 0 | 0 |
Period 3 (7-week Therapy Period)
| Stage | Duloxetine 60 mg With Food | Duloxetine 60 mg Without Food | Duloxetine 30 mg With Food | Duloxetine 30 mg Without Food |
|---|---|---|---|---|
| STARTED | 56 | 59 | 59 | 61 |
| COMPLETED | 26 | 36 | 39 | 36 |
| NOT COMPLETED | 30 | 23 | 20 | 25 |
Reasons for not completing this period:
| Reason | Duloxetine 60 mg With Food | Duloxetine 60 mg Without Food | Duloxetine 30 mg With Food | Duloxetine 30 mg Without Food |
|---|---|---|---|---|
| Adverse Event | 17 | 15 | 10 | 12 |
| Death | 0 | 0 | 1 | 0 |
| Lost to Follow-up | 0 | 0 | 1 | 0 |
| Protocol Violation | 10 | 4 | 3 | 11 |
| Withdrawal by Subject | 2 | 2 | 4 | 2 |
| Lack of Efficacy | 1 | 2 | 1 | 0 |
Analysis population for the primary outcome, as defined in the registry: The safety population included all participants who took at least 1 study drug dose analysed according to the drug dose actually taken.
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 | Duloxetine 60 mg With Food | Duloxetine 60 mg Without Food | Duloxetine 30 mg With Food | Duloxetine 30 mg Without Food | Total |
|---|---|---|---|---|---|---|
| Participants analysed | 59 | 63 | 63 | 64 | 249 | |
| Age, Continuous | years | 44.65 (31.58 to 53.73) | 47.90 (38.65 to 58.90) | 49.94 (40.04 to 62.24) | 44.65 (32.84 to 56.92) | 46.81 (34.78 to 59.25) |
| Sex: Female, Male: Female | Participants | 43 | 47 | 47 | 40 | 177 |
| Sex: Female, Male: Male | Participants | 16 | 16 | 16 | 24 | 72 |
| Race/Ethnicity, Customized: Korean | participants | 59 | 63 | 63 | 64 | 249 |
| Region of Enrollment: Korea, Republic of | participants | 59 | 63 | 63 | 64 | 249 |
| Body Mass Index (BMI) | kilogram per square meter (kg/m^2) | 24.56 (SD 4.034) | 23.28 (SD 3.133) | 23.06 (SD 3.430) | 23.84 (SD 3.612) | 23.67 (SD 3.584) |
| Previously diagnosed with major depressive disorder (MDD) | participants | 59 | 63 | 63 | 64 | 249 |
| Duration since first major depressive disorder (MDD) episode | years | 1.47 (0.52 to 4.29) | 2.65 (0.77 to 5.32) | 2.52 (0.48 to 9.12) | 1.29 (0.52 to 4.40) | 1.92 (0.54 to 5.55) |
| Age at first major depressive disorder (MDD) episode | years | 41.21 (29.00 to 51.00) | 43.33 (34.67 to 53.00) | 43.86 (27.00 to 59.60) | 41.58 (29.00 to 54.00) | 42.52 (30.00 to 54.00) |
| Number of previous MDD episodes/exacerbations in the last 24 months | number of episodes in last 24 months | 1.0 (0.0 to 1.0) | 1.0 (0.0 to 1.0) | 1.0 (0.0 to 1.0) | 1.0 (0.0 to 1.0) | 1.0 (0.0 to 1.0) |
| Received previous therapy for current episode: yes | participants | 18 | 25 | 18 | 26 | 87 |
| Received previous therapy for current episode: no | participants | 41 | 38 | 45 | 38 | 162 |
| 17-item Hamilton Depression Rating Scale (HAMD-17) Total Score | units on a scale | 21.0 (SD 5.18) | 22.3 (SD 5.69) | 22.9 (SD 5.45) | 20.3 (SD 5.23) | 21.6 (SD 5.46) |
| Clinical Global Impression of Severity (CGI-S) Score | units on a scale | 4.4 (4.0 to 5.0) | 4.5 (4.0 to 5.0) | 4.7 (4.0 to 5.0) | 4.2 (4.0 to 5.0) | 4.4 (4.0 to 5.0) |
| Association for Methodology and Documentation in Psychiatry (AMDP-5) Item 112 (Nausea) Score | units on a scale | 0.3 (SD 0.72) | 0.2 (SD 0.59) | 0.2 (SD 0.61) | 0.1 (SD 0.38) | 0.2 (SD 0.58) |
6. Primary Endpoint
The registry lists the primary outcome as Mean Maximum Nausea Severity, Association for Methodology and Documentation in Psychiatry (AMDP-5) Adverse Event (AE) Scale Item 112 (Nausea), measured over 1 week and 8 weeks.
| Element | Registry entry |
|---|---|
| Result type | Mean |
| Measure of spread | 95% Confidence Interval |
| Unit | units on a scale |
| Endpoint classification | Continuous Endpoint |
The primary measure was mean maximum nausea severity on item 112 of the AMDP-5 adverse event scale. The scale runs from 0 to 3: 0 means nausea was not present, 1 means mild, 2 means moderate, and 3 means severe. The endpoint description says nausea severity was measured during treatment from Week 0-8, with timeframes at 1 week and 8 weeks.
Nausea was selected because the study purpose was to assess nausea severity in response to different duloxetine dosing strategies. A maximum severity measure is intended to capture the highest recorded level of the symptom over the stated assessment period, rather than only whether nausea occurred at a single moment. The registry identifies the summary as a mean and specifies a 95% confidence interval as the dispersion measure.
As a general statistical point, the scale categories are ordered levels of severity. A mean summarizes scores numerically, but it does not show how many participants experienced each category. Category-level results could add context, but they are not reported in the registry record. Nor are the means, confidence intervals, or between-group estimates reported, so the degree of nausea severity in each group and the precision of comparisons cannot be assessed from the registry record.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record identifies mean maximum nausea severity at 1 week and 8 weeks as the primary outcome, but it does not report the numerical group means, confidence intervals, or comparisons between the four strategies. It therefore does not support a numerical description of how nausea severity varied across groups. The record also does not report a p-value or other inferential result for the primary outcome.
The specified 95% confidence interval would ordinarily describe the uncertainty around an estimated mean or contrast. In general, a narrower interval indicates greater precision, while a wider interval indicates more uncertainty. The interval’s interpretation depends on what quantity it surrounds: a group mean describes uncertainty in that mean, while a between-group contrast describes uncertainty in the estimated difference. Since no primary outcome estimates or intervals are reported, neither the magnitude nor precision of the study’s nausea findings can be evaluated.
The adverse event counts are not a substitute for the primary continuous outcome. Counts of participants affected by serious or other adverse events summarize different aspects of safety and do not give the mean maximum nausea score. The record reports serious adverse events affecting 1 participant in the 30 mg with-food group, 1 in the 60 mg with-food group, 2 in the 30 mg without-food group, and 0 in the 60 mg without-food group. It separately reports other adverse events affecting 52, 55, 57, and 58 participants, respectively. These figures do not provide the missing nausea means or a direct statistical comparison of the primary endpoint.
It would also be inappropriate to interpret the absence of a reported comparison as evidence that the regimens had identical results. Without estimates and uncertainty intervals, the registry record cannot distinguish a small difference estimated precisely from a larger difference estimated imprecisely, or establish whether any observed difference was compatible with chance variation. A non-significant result cannot be inferred because no significance test result is reported.
The treatment strategies combine dose and food instructions, and two strategies change dose after 1 week. A comparison between groups therefore concerns the assigned regimens as a whole. Isolating a dose effect, a food effect, or an interaction between dose and food would require specified contrasts and an analysis; those details are not reported in the registry record. The available record supports describing the question and design, but not claiming that one strategy produced more or less nausea than another.
9. Statistical Methods Explained
The registry describes a randomized, parallel-group design and names a continuous primary measure summarized as a mean with a 95% confidence interval. It does not report the statistical analysis methods used for the primary outcome, the model or test used, or numerical estimates. The statistical explanations below are general and do not describe an analysis known to have been performed in this study.
What does a mean summarize?
A mean combines participants’ numerical scores into one measure of the group’s average score. Here, the score represents nausea severity on an ordered scale from 0 to 3. A group mean can be useful for comparing overall levels, but it does not show the full distribution: for example, it does not by itself indicate how many participants reported no nausea or severe nausea. The registry’s definition of the endpoint is mean maximum nausea severity, rather than a count of participants with any nausea.
Because the scale consists of ordered categories, the distance between adjacent categories is not necessarily identical in lived experience. Treating the scores numerically and calculating a mean is a summary choice, not proof that the categories are equally spaced. It can be informative to interpret a mean alongside category frequencies, but those frequencies are not reported for the primary outcome in the registry record.
What does a confidence interval communicate?
In general, a confidence interval gives a range of estimates compatible with the data and the method used to construct it. It conveys precision: a more compact interval represents more precise estimation than a broader interval. It does not mean that a fixed proportion of individual participants’ scores fall inside the interval, and it does not establish that a treatment effect is important. For a comparison, the interval must surround a contrast such as a difference between groups; intervals around separate group means are not, by themselves, a direct test of the difference.
The record specifies 95% confidence intervals for the primary measure but provides no numerical intervals. It is not reported whether intervals were calculated separately for each group, for comparisons between groups, or both. Their construction method is also not reported. Consequently, no conclusion about the primary endpoint’s precision can be drawn from the recorded endpoint description alone.
What does randomization contribute?
As a general principle, random allocation can help balance measured and unmeasured participant characteristics across groups on average. It supports comparisons between assigned strategies, but it does not guarantee that groups are identical, especially in an individual trial. Interpretation also depends on how randomization was carried out, whether assigned groups were maintained in the analysis, and how missing outcomes were handled. The registry record identifies the study as randomized but does not report the randomization procedure, allocation concealment, or the primary outcome analysis population beyond its safety-population description.
The study is described as open-label. In general, lack of masking can matter more for outcomes that involve subjective reporting than for outcomes measured by an instrument independent of participant or assessor judgment. That consideration does not show that the nausea ratings were affected; it identifies a possible source of measurement influence that should be considered when reading the design.
How should the planning illustration be read?
The planning illustration is distinct from an observed trial result and from the trial’s original design justification. Its stated assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is a stated fraction of the observed effect. Those assumptions describe an illustration, not a statistical plan confirmed for this trial. The registry record does not report the trial’s own sample-size justification or the observed primary outcome effect needed to interpret a numerical planning estimate.
Such planning exercises ask how many participants per arm might be needed under particular assumptions to detect a chosen effect with a selected level of power. They are conditional on the effect size, variability, analysis, and allocation assumptions used. They are not a measure of the probability that the completed study found a true effect, and they should never be described as observed power. In this case, no numerical planning estimate or effect value is reported in the record being summarized.
Why do distributional assumptions matter?
General statistical methods rely on assumptions about the data and the way observations relate to one another. A normal approximation may be used for some mean-based comparisons, but its suitability depends on the outcome distribution, sample size, and analysis details. For count data, over-dispersion means that variation exceeds what a chosen count model expects; if ignored, uncertainty can be understated. This is a general concept and does not imply that the trial used a count model or had an over-dispersion problem. The registry record does not describe such modeling for nausea or adverse events.
Overall, the registry identifies the endpoint and design but does not give enough information to reproduce or evaluate a specific inferential analysis. The analysis model, handling of repeated assessments, adjustment for baseline measurements, treatment of missing data, and multiplicity strategy are not reported in the registry record.
10. Multiplicity, Missing Data and Analysis Populations
The study includes multiple outcomes and assessment times. The registry lists 16 outcome measures, including nausea, gastric events, common adverse events, and several HAMD-17 scores or subscales. The primary nausea endpoint has timeframes at 1 week and 8 weeks. The record does not report how many hypotheses were tested or how the primary outcome’s time points were handled analytically; it records no analyses.
When multiple outcomes, time points, or group contrasts are tested, the chance of at least one apparently unusual result can increase. General approaches include designating a primary comparison in advance, adjusting significance thresholds, or treating secondary analyses as exploratory. Which approach was used in this study is not reported in the registry record. No multiplicity adjustment or hierarchy of tests should be inferred from the outcome list alone.
The primary outcome population is described as the safety population: participants who took at least 1 study drug dose, analysed according to the drug dose actually taken. This is not necessarily the same as an analysis based on randomized assignment. The registry record does not report the number analysed for the primary outcome, the handling of participants who changed or stopped treatment, or whether any intention-to-treat analysis was performed.
Missing outcome counts, reasons for missingness, and the methods used to address missing data are not reported in the registry record. This matters because an analysis using only observed assessments can be influenced if missingness is related to nausea or treatment experience. No assumptions about missing outcomes, imputation, or sensitivity analyses can be made from the reported information.
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 |
|---|---|---|---|
| Duloxetine 30 mg With Food | 1 | 63 | 1.6% |
| Duloxetine 60 mg With Food | 1 | 59 | 1.7% |
| Duloxetine 30 mg Without Food | 2 | 64 | 3.1% |
| Duloxetine 60 mg Without Food | 0 | 63 | 0.0% |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 5% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Duloxetine 30 mg With Food | 52 of 63 |
| Duloxetine 60 mg With Food | 55 of 59 |
| Duloxetine 30 mg Without Food | 57 of 64 |
| Duloxetine 60 mg Without Food | 58 of 63 |
Most frequent serious adverse event terms
| Event term | Organ system | Duloxetine 30 mg With Food | Duloxetine 60 mg With Food | Duloxetine 30 mg Without Food | Duloxetine 60 mg Without Food |
|---|---|---|---|---|---|
| Diarrhoea | Gastrointestinal disorders | 0 of 63 | 1 of 59 | 0 of 64 | 0 of 63 |
| Pelvic inflammatory disease | Infections and infestations | 0 of 63 | 1 of 59 | 0 of 64 | 0 of 63 |
| Muscle injury | Injury, poisoning and procedural complications | 0 of 63 | 0 of 59 | 1 of 64 | 0 of 63 |
| Road traffic accident | Injury, poisoning and procedural complications | 0 of 63 | 0 of 59 | 1 of 64 | 0 of 63 |
| Lipoma | Neoplasms benign, malignant and unspecified (incl cysts and polyps) | 0 of 63 | 0 of 59 | 1 of 64 | 0 of 63 |
| Completed suicide | Psychiatric disorders | 1 of 63 | 0 of 59 | 0 of 64 | 0 of 63 |
| Ovarian cyst | Reproductive system and breast disorders | 0 of 63 | 1 of 59 | 0 of 64 | 0 of 63 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Duloxetine 30 mg With Food | Duloxetine 60 mg With Food | Duloxetine 30 mg Without Food | Duloxetine 60 mg Without Food |
|---|---|---|---|---|---|
| Nausea | Gastrointestinal disorders | 39 of 63 | 41 of 59 | 38 of 64 | 39 of 63 |
| Decreased appetite | Metabolism and nutrition disorders | 11 of 63 | 11 of 59 | 13 of 64 | 11 of 63 |
| Headache | Nervous system disorders | 11 of 63 | 10 of 59 | 14 of 64 | 8 of 63 |
| Dry mouth | Gastrointestinal disorders | 10 of 63 | 10 of 59 | 10 of 64 | 12 of 63 |
| Constipation | Gastrointestinal disorders | 12 of 63 | 9 of 59 | 9 of 64 | 10 of 63 |
| Dizziness | Nervous system disorders | 11 of 63 | 8 of 59 | 10 of 64 | 7 of 63 |
| Vomiting | Gastrointestinal disorders | 2 of 63 | 8 of 59 | 8 of 64 | 10 of 63 |
| Insomnia | Psychiatric disorders | 8 of 63 | 8 of 59 | 6 of 64 | 5 of 63 |
| Sedation | Nervous system disorders | 5 of 63 | 6 of 59 | 5 of 64 | 6 of 63 |
| Abdominal discomfort | Gastrointestinal disorders | 3 of 63 | 6 of 59 | 7 of 64 | 4 of 63 |
Serious events with the highest arm-level rate: diarrhoea (1.7%), pelvic inflammatory disease (1.7%), ovarian cyst (1.7%), completed suicide (1.6%), muscle injury (1.6%).
The registry reports serious and other adverse event counts by dosing group. For serious adverse events, the numbers at risk and affected were 63 and 1 for duloxetine 30 mg with food; 59 and 1 for duloxetine 60 mg with food; 64 and 2 for duloxetine 30 mg without food; and 63 and 0 for duloxetine 60 mg without food. For other adverse events, the corresponding at-risk and affected counts were 63 and 52, 59 and 55, 64 and 57, and 63 and 58.
These are counts of participants, not estimates of the primary mean nausea score. They should be read separately from the primary endpoint and do not establish a causal difference between regimens. The registry also lists Diarrhoea, Pelvic inflammatory disease, and Ovarian cyst at 1.7% each, and Completed suicide, Muscle injury, and Road traffic accident at 1.6% each. The listed terms do not specify an arm attribution or denominator in the registry record. Deaths are not reported in the registry record.
As a general caution, event counts can be affected by how events are defined, collected, and coded. The record’s counts alone do not describe timing, duration, severity beyond the serious classification, or whether an event was judged related to treatment.
12. What the Result Does, and Does Not, Mean
This trial addresses a focused statistical question: how nausea severity was recorded under four duloxetine dosing strategies in Korean patients with major depressive disorder. Its randomized parallel-group design is relevant to comparing the assigned regimens, and its outcome definition specifies a severity scale and assessment times. The registry record does not report the primary outcome values or comparative estimates, so it does not show the size or direction of any difference in nausea severity.
The missing numerical results are important to interpretation. Without group means and confidence intervals, readers cannot assess how high the recorded severity was, how much the groups differed, or how precisely the endpoint was estimated. The adverse event counts offer separate safety information, but they cannot fill in those missing continuous outcome results. Nor can an unreported p-value be treated as evidence for or against a difference.
The design also compares regimens rather than isolated ingredients of a regimen. The groups differed in food instruction, while two groups used 30 mg for 1 week before changing to 60 mg for 7 weeks. Any valid contrast would therefore need to be interpreted in light of these schedules. The registry record does not report whether the analysis separated dose and food effects.
More generally, a registry description is not a full account of every analytic decision or every result. The record does not report the statistical model, missing-data methods, multiplicity strategy, or primary outcome estimates. These absences limit what can be concluded from this record; they do not establish that the study had no analyses or that the strategies were equivalent. No clinical recommendation follows from the information.
13. Why This Trial Matters Statistically
This study is statistically interesting because its four arms vary both food instruction and dosing schedule while focusing on nausea severity. The comparison is not simply between two fixed doses: the 30 mg strategies use 30 mg for 1 week and then 60 mg for 7 weeks, whereas the 60 mg strategies use 60 mg for 8 weeks. That structure matters when defining what a between-group estimate would mean.
The endpoint also illustrates a common challenge in clinical research: a symptom is recorded on an ordered severity scale, then summarized as a mean. The mean can provide a compact group-level summary, while leaving unanswered how scores are distributed across severity categories. A careful interpretation would consider the endpoint definition, the summary chosen, and the uncertainty around each estimate.
Open-label assessment is another relevant design feature. As a general matter, awareness of treatment assignment can influence reporting of subjective symptoms, although the registry record does not show that this occurred. Randomization and clear outcome measurement remain important, but neither removes the need to inspect how the outcome was collected and analyzed.
Finally, this record illustrates the difference between a detailed study question and an interpretable result. It identifies the population, four strategies, outcome scale, and assessment times, but does not report numerical primary outcome results or the analysis methods. Without those estimates and their uncertainty, the statistical comparison itself cannot be judged. That distinction is useful when reading registry records: design information can define the intended question without supplying the evidence needed to evaluate its answer.
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 limitation for interpretation is that the registry record does not report numerical results for mean maximum nausea severity. Group estimates, confidence intervals, and between-group comparisons are absent, so the main study question cannot be answered quantitatively from this record. No primary outcome p-value or analysis method is reported either.
The open-label design is relevant because nausea severity is a symptom rating. In general, knowledge of treatment assignment may influence symptom reporting or assessment, but the record does not establish that this affected the results. The randomization method and allocation concealment are also not reported, which limits assessment of how assignment was implemented.
The dosing regimens differ in more than one respect. Food instruction varies, and the groups called 30 mg change to 60 mg after 1 week. Thus, the comparison concerns combined dosing strategies rather than a sustained 30 mg versus 60 mg contrast. The registry does not report which contrasts were planned or whether dose and food effects were analyzed separately.
The endpoint is assessed at 1 week and 8 weeks, but the record does not report the numerical results at either time or how repeated assessments were analyzed. Missing data counts and methods are also not reported. The primary population is a safety population of participants who took at least 1 dose and were analyzed according to dose actually taken; other analysis populations are not described.
The study concerns Korean patients with major depressive disorder and treatment over 8 weeks. The record does not provide enough detail to assess how broadly its findings would apply beyond that population and duration. Since the primary outcome results are not reported, these limitations constrain interpretation rather than establish any particular outcome.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: A Duloxetine Dosing Strategy Study in Korean Patients With Major Depressive Disorder — NCT00960986.
- Registry record last updated 2014-12-31; status last verified 2014-12.
- Results first posted 2012-04-17.
- 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 registry record describes a completed, randomized, open-label, parallel-group phase 4 study of duloxetine dosing strategies in Korean patients with major depressive disorder. Enrollment was 249. The four regimens differed by food instruction and by whether participants took 60 mg throughout 8 weeks or started at 30 mg for 1 week before taking 60 mg for 7 weeks.
The primary outcome was mean maximum nausea severity on AMDP-5 adverse event scale item 112, scored from 0 to 3 and assessed at 1 week and 8 weeks. The primary population was the safety population, consisting of participants who took at least 1 dose and analyzed according to the dose actually taken. The registry reports adverse event counts but does not report numerical primary outcome results, comparisons, or the statistical analysis methods. The study start was 2009-08, primary completion was 2011-02, and status is completed.