Efficacy of Shinabro in Hand Osteoarthritis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Efficacy of Shinabro in Hand Osteoarthritis (NCT01910116), a randomized phase 2, phase 3 study. The primary outcome is AUSCAN Pain Change at 4 Weeks From Baseline.

2. Clinical Question

The trial asks whether change in hand-osteoarthritis pain over four weeks differs between participants assigned to Shinbaro and those assigned to placebo. Pain was measured using the AUSCAN scale, which ranges from 0 (no pain) to 100 (worst possible pain). The recorded change is the score at four weeks minus the score at baseline, so negative values indicate improvement from baseline and positive values indicate deterioration.

The primary endpoint is measured at four weeks even though the assigned regimens are described for 12 weeks. The record also lists pain and other outcomes at later time points, as well as stiffness, function, and responder outcomes. Those measures provide additional views of the trial, but they do not change the definition of its primary question.

The registry identifies the study as randomized, double-blinded, placebo-controlled, and multicenter. It does not report the detailed allocation procedure, the identity of each masked role, or the full clinical criteria used to define the study population. The trial record therefore establishes the broad comparison and endpoint, but not every detail needed to reconstruct how participants were selected or how treatment delivery was managed.

3. Trial Design

This was a randomized, parallel-group comparison of Shinbaro with placebo. The registry describes the masking as quadruple and the study as double-blinded in its official title and summary. It identifies the trial as phase 2 and phase 3, and as an investigator-initiated study sponsored by Seoul National University Hospital. The registry lists a total enrollment of 220.

The intervention descriptions specify 2 capsules twice daily for 12 weeks in each group. For Shinbaro, the arm description also states 300mg. The placebo arm is described as receiving placebo capsules on the same schedule. These descriptions establish the planned treatment regimens, but the registry record does not report adherence, treatment discontinuation, or the extent to which participants completed the assigned regimen.

The study started in 2013-09, reached primary completion in 2014-11, and is listed as completed. The primary endpoint is change in AUSCAN pain from baseline to four weeks. Other posted outcomes include pain at later time points, stiffness, function, and an OMERACT-OARSI responder count. The record identifies multiple secondary outcome measures, but the complete protocol and the full schedule of all assessments are not reported in the registry record.

The outcome groups listed for the primary comparison contain 109 participants in the Shinbaro arm and 106 in the placebo arm. These are the group sizes shown with the posted primary endpoint, while the registry separately gives total enrollment as 220. The record does not explain the relationship between enrollment and the analysis groups, or specify whether the primary analysis used all randomized participants, participants with observed endpoint data, or another analysis population.

Details such as the randomization sequence generation, concealment procedure, stratification, site-specific procedures, and handling of protocol deviations are not reported in the registry record. Nor is a sample-size justification or a full statistical analysis plan reported. These omissions limit how completely the design can be independently reconstructed from the registry entry alone.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

Reasons for not completing this period:

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.

Population: 1 patients in the Shinbaro group (n=110) and 3 in the placebo group (n=110) did not receive the allocated intervention as they withdrew immediately after randomization. Also, information on 1 patient in the placebo group was not complete leading to drop out.

6. Primary Endpoint

The registry lists the primary outcome as AUSCAN Pain Change at 4 Weeks From Baseline, measured over Baseline and 4 weeks.

The primary endpoint is the change in AUSCAN pain score from baseline to four weeks. The registry defines the change as the four-week score minus the baseline score. AUSCAN pain is measured on a scale from 0 (no pain) to 100 (worst possible pain), and the direction is explicitly stated: a negative change means improvement from baseline, while a positive change means deterioration.

A change score focuses on how a participant's recorded pain score moved over the stated interval, rather than on the follow-up score alone. In general, a baseline-to-follow-up change measure can make the comparison relevant to the amount of movement during the assessment period. That general interpretation does not establish how baseline scores were handled in the analysis; the detailed analysis plan is not reported in the registry record.

The endpoint is summarized using medians and interquartile ranges. A median describes the middle of the observed distribution, while the interquartile range describes the spread of its central portion. These summaries are useful when the analysis is based on ranks or when a distribution is not well represented by a mean and standard deviation. The registry reports a Wilcoxon (Mann-Whitney) comparison for the two groups.

The endpoint concerns pain change at four weeks. It does not, by itself, describe later pain, stiffness, function, overall response, or outcomes beyond the assessment period. It also does not state how every participant felt or whether individual changes were clinically important. The registry record does not report a threshold for a clinically important difference in this endpoint.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint is a change score, defined as the four-week AUSCAN pain score minus the baseline score. With this definition, a negative number points toward a lower pain score at follow-up, while a positive number points toward a higher score. The registry summarizes the arm-level changes with medians and interquartile ranges rather than means and standard deviations.

What does the median show?

The median is the middle value after observations are ordered. It is a summary of the centre of a distribution, not a statement that every participant had that value. An interquartile range describes the interval containing the central half of the ordered observations. It conveys spread, but it does not show the full shape of the distribution or how individual observations are paired across time. The posted primary values are arm summaries; they do not provide individual participant trajectories.

Why use a Wilcoxon (Mann-Whitney) test?

The registry reports a Wilcoxon (Mann-Whitney) test for the primary two-group comparison. In general, this is a rank-based method: it orders observations and compares their relative positions between groups, rather than relying directly on a comparison of arithmetic means. It can be useful when an analysis is framed around distributions or when assumptions associated with a mean-based comparison are not appropriate. A rank-based test does not automatically test only a difference in medians. Interpreting it as a test of medians requires additional assumptions about the shapes of the group distributions, and those assumptions are not documented in the registry record.

The test's reported p-value is 0.014. A p-value summarizes how incompatible the observed test result is with a null model of no group difference, assuming that model and the analysis assumptions hold. It does not measure the probability that a treatment effect is absent, and it does not quantify the magnitude or practical importance of an effect. The primary result does not include a posted confidence interval for a between-group contrast, so a direct range of plausible contrast values is not available from the primary result as recorded.

What does a confidence interval mean?

In general, a confidence interval communicates uncertainty around an estimated effect. Its width reflects precision under the method used: a narrower interval indicates a more precise estimate, while a wider interval indicates greater uncertainty. The secondary responder calculation has a confidence interval for a risk difference, but the primary result does not report a corresponding between-group interval. The arm-level interquartile bounds are not a confidence interval for the treatment contrast; they summarize the observed spread within each arm.

How should the responder comparison be read?

The secondary OMERACT-OARSI responder outcome is posted as a number of participants, and the recalculated comparison expresses the group contrast as a risk difference. A risk difference is the difference between the proportions meeting the responder definition in the two groups. In general, it is an absolute measure, unlike a risk ratio or odds ratio, which compare groups on relative scales. The recalculated estimate is 0.127, with a confidence interval from -0.004 to 0.259 and p = 0.06. Since that interval includes zero, it is compatible with no difference as well as with differences in either direction under the stated calculation. This is an unadjusted recalculation, not a substitute for a sponsor-specified analysis.

The registry reports a chi-squared test for the responder comparison with p = 0.060. The recalculated unadjusted p-value is 0.06. These values are presented at different displayed precision, and the calculation is not adjusted for other endpoints or comparisons. The trial's own reported analysis is the primary reference for its planned inferential result; where no discrepancy is flagged, the recalculation is best viewed as a transparent supplemental summary from the posted group counts.

What are planning illustrations?

The sample-size section presents planning illustrations based on fractions of the observed responder effect. It reports participants per arm for 80% power and participants per arm for 90% power, under the listed assumptions of a two-sided significance level of 0.05, equal allocation, and a normal approximation. These are illustrations derived from the observed effect, not the trial's own sample-size justification and not observed power. Planning for a smaller effect generally requires more participants than planning for a larger effect, all else equal. That general principle explains the purpose of the illustrations but does not establish which effect the investigators expected when designing the study.

The registry record does not report the original sample-size calculation, its assumed effect, or any interim analysis plan. It also does not report enough detail to determine how missing observations were handled, which analysis population was used, or whether the primary analysis adjusted for baseline values or study site. These are important components of statistical methods, and their absence means the reported summaries cannot fully reconstruct the analysis.

10. Analyses Reported by the Sponsor

The registry's own primary analysis is the Wilcoxon (Mann-Whitney) comparison, reported with p = 0.014. That sponsor-reported result should be read as the primary inferential analysis recorded for the four-week AUSCAN pain endpoint. The posted group medians and interquartile bounds describe each arm, but they are not themselves an unadjusted recalculation of the between-arm test.

For the secondary responder outcome, the registry reports a chi-squared test with p = 0.060. The unadjusted recalculation from the posted counts gives p = 0.06 and a risk difference of 0.127, with a confidence interval from -0.004 to 0.259. The small difference in displayed precision between p = 0.060 and p = 0.06 is consistent with rounding. The record does not flag a discrepancy for these results.

A recalculation from summary counts is useful as a transparent check, but it may not reproduce every feature of a planned analysis, such as covariate adjustment or handling of missing observations. Those details are not reported in the registry record. The sponsor's reported analysis therefore remains the principal reference for what the trial reported, while the recalculation is a supplemental unadjusted view of the posted responder counts.

11. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary endpoint and many secondary outcome measures, including repeated assessments of pain, stiffness, and function. Testing several outcomes and time points creates a multiplicity issue. In general, when multiple hypotheses are tested, the chance of at least one small p-value arising by chance can be greater than for a single test. Common approaches include adjusting significance thresholds or defining a testing hierarchy in advance. The registry record does not report whether any such procedure was used, or which secondary outcomes were intended to support confirmatory claims.

The record reports 35 secondary analyses and 36 outcome measures, but does not provide a complete statistical analysis plan. Without that plan, it is not possible to determine how the analyses were grouped, whether tests were prespecified, whether the primary outcome was the only confirmatory test, or whether adjustments were applied. The secondary responder p-value should therefore be treated as unadjusted in the recalculation and interpreted cautiously.

Missing data are another important consideration for a change-from-baseline endpoint. In general, participants may lack a follow-up measurement, and methods for handling those missing observations can affect both the estimate and its uncertainty. The registry record does not report how missing primary or secondary measurements were handled, whether any imputation was used, or what assumptions underlay the analysis.

The analysis population is also not identified in sufficient detail. The registry gives total enrollment of 220 and lists 109 participants in the Shinbaro group and 106 in the placebo group for the primary endpoint. It does not explain whether these numbers represent randomized participants, participants with observed outcomes, or another population, nor does it report how protocol deviations were treated. Without this information, the reader cannot establish whether the analysis followed an intention-to-treat approach or another rule.

These reporting gaps do not invalidate the posted results, but they limit independent assessment of their robustness. The trial protocol, analysis plan, missing-data details, and analysis-population definitions are not reported in the registry record.

12. 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.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: fracture (0.9%), surgery (0.9%), skin rash (0.9%), liver function abnormality (0.9%).

The posted safety summaries list serious adverse events for 1 of 109 participants in the Shinbaro group and 4 of 106 participants in the placebo group. The registry also lists other adverse events for 55 of 109 participants in the Shinbaro group and 45 of 106 participants in the placebo group. These are counts as recorded in the registry; they do not establish whether an event was caused by the assigned treatment.

The listed serious adverse event terms include fracture, surgery, skin rash, and liver function abnormality, each shown as 0.9%. The record does not establish from these terms alone whether they were judged treatment-related, how severity was classified, or whether one participant could contribute more than one event. Deaths are not reported in the registry record.

Safety counts should be interpreted alongside the study's duration, follow-up, and reporting definitions. The assigned regimens are described for 12 weeks, but the registry summary does not provide detailed event timing, exposure-adjusted rates, or a full account of event ascertainment. Counts from this trial describe what was posted for its groups; they do not by themselves establish comparative safety beyond the recorded study.

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

The primary result describes a comparison of four-week AUSCAN pain change between randomized groups. The group medians were -9.0 for Shinbaro and -2.2 for placebo, with negative values defined as improvement from baseline. The registry's Wilcoxon (Mann-Whitney) test reports p = 0.014. Taken together, these show a difference in the rank-based comparison as reported in the registry, alongside arm-level summaries that both have negative medians.

These figures do not establish the size of a direct between-group difference in the primary endpoint, because the registry does not post a primary treatment contrast with a confidence interval. A p-value alone does not communicate the magnitude of an effect or its clinical importance. The record also does not report a threshold for clinically important change, so the endpoint summaries cannot be categorized against such a threshold.

The result applies to the measured endpoint and assessment time: AUSCAN pain change from baseline to four weeks. It does not directly describe later pain, stiffness, function, the full course of symptoms, or outcomes outside the recorded study period. The listed regimen lasts 12 weeks, but that does not change the four-week timing of the primary endpoint.

The trial was randomized and placebo-controlled, but important interpretive details remain unavailable in the registry record, including the analysis population, missing-data methods, and multiplicity plan. The secondary responder comparison is not a replacement for the primary analysis, and the other secondary results cannot be summarized as a coherent pattern from the information.

Accordingly, the posted results are best read as trial-specific statistical evidence about a defined outcome, not as a complete account of benefit, harm, or longer-term experience. They do not, by themselves, establish what an individual participant would experience or support medical advice.

14. Why This Trial Matters Statistically

This trial is statistically instructive because it places a rank-based primary comparison beside arm-level summaries expressed as medians and interquartile ranges. The distinction matters: the group medians describe the centre of each arm, while the Mann-Whitney test compares ranked observations. A p-value from that test should not be mistaken for a direct estimate of the difference between the medians.

The endpoint is a change score with a clear direction: negative values indicate improvement from baseline and positive values indicate deterioration. The posted summaries therefore require attention to the endpoint's definition, not just the sign of a number. The different information conveyed by medians, interquartile ranges, and a p-value is a useful reminder that no single statistic describes the whole result.

The secondary responder comparison also illustrates the difference between a trial-reported test and an unadjusted recalculation. Its risk-difference estimate and confidence interval show uncertainty in an absolute contrast, while the registry's chi-squared p-value records the sponsor's reported test. The fact that multiple outcomes and time points are listed makes the missing multiplicity plan particularly relevant when interpreting secondary analyses.

Finally, the registry provides a useful example of the limits of abbreviated reporting. It identifies randomization, masking, treatment schedules, outcomes, and group summaries, but does not report key details about missing observations, analysis populations, or the original sample-size rationale. Statistical interpretation depends not only on calculating a test, but also on understanding how the data entered that test and what claims the analysis was designed to support.

15. 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.

Several limitations are specific to what this registry record reports. The trial is described as randomized and quadruple-masked, but details of sequence generation, allocation concealment, and the roles covered by masking are not reported in the registry record. The official title and summary also use the term double-blinded; the record does not explain the relationship between these masking descriptions.

The primary endpoint summary gives arm-level medians and interquartile bounds but does not provide a between-group effect estimate or confidence interval. The reported rank-based p-value therefore cannot, on its own, convey the magnitude or precision of a direct primary contrast. The record also does not state whether the test was adjusted for baseline values, site, or other factors.

The total enrollment is reported as 220, while the primary outcome groups are listed as 109 and 106. The registry does not explain the relationship between these figures or identify the analysis population. It does not report missing-data methods, adherence, discontinuations, or protocol deviations. These omissions limit assessment of how representative the analyzed participants were of everyone randomized.

The record lists many secondary outcomes and analyses but does not report a multiplicity strategy or a hierarchy of endpoints. The numerical summary does not establish the overall direction of the secondary findings. Consequently, secondary results cannot be interpreted as a unified confirmation of the primary endpoint from this record alone.

The sample-size planning illustrations are based on the observed responder effect and are not the trial's original design justification. The original sample-size assumptions and any interim analysis plan are not reported in the registry record. Finally, the registry's safety summary provides counts and selected event terms without detailed timing, causality assessment, or a full account of event definitions. These limitations concern the available reporting and should not be mistaken for proof that an unreported procedure was or was not performed.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Efficacy of Shinabro in Hand Osteoarthritis — NCT01910116.
• Registry record last updated 2015-12-17; status last verified 2015-08.
• Results first posted 2015-12-17.
• 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.

19. Record Summary

NCT01910116 is a completed randomized, parallel-group, placebo-controlled trial of Shinbaro for hand osteoarthritis. The registry lists phase 2 and phase 3, quadruple masking, sponsor Seoul National University Hospital, and total enrollment of 220. The treatment arms are described as receiving Shinbaro or placebo twice daily for 12 weeks.

The primary endpoint is change in AUSCAN pain from baseline to four weeks. The registry reports medians of -9.0 for Shinbaro and -2.2 for placebo, with interquartile bounds, and a Wilcoxon (Mann-Whitney) p-value of 0.014. Negative changes are defined as improvement from baseline. The primary result does not include a between-group confidence interval.

Secondary measures cover pain, stiffness, function, and responder status. The registry also posts safety counts, including serious adverse events in both groups. Key details on multiplicity, missing data, analysis populations, and the original sample-size rationale are not reported in the registry record.