A Study to Evaluate the Efficacy of Sativex in Relieving Symptoms of Spasticity Due to Multiple Sclerosis: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study to Evaluate the Efficacy of Sativex in Relieving Symptoms of Spasticity Due to Multiple Sclerosis (NCT01599234), a randomized phase 3 study. The primary outcome is Change From Baseline in Mean Spasticity 0-10 Numerical Rating Scale (NRS) Score During the Last 14 Day of Treatment (End of Treatment).

2. Clinical Question

The study title and summary identify relief of spasticity symptoms in multiple sclerosis as the clinical topic. The primary measure was recorded by participants at bedtime each day, asking them to rate their spasticity over the last 24 hours. The registry describes the primary timeframe as 0-15 weeks and identifies the endpoint as change from baseline at the end of treatment. The main comparison therefore concerns average change in a participant-reported symptom score, not a count of participants who reached a specific improvement threshold.

The registry also lists secondary measures relating to spasticity improvement thresholds, muscle tone, sleep quality, walking time, daily activities, carer impression, and adverse events. These outcomes provide additional perspectives, but they do not replace the primary question. The record does not state a minimum change on the spasticity scale that would be considered clinically important, so the clinical importance of a particular score difference cannot be determined from the registry alone.

3. Trial Design

This was a phase 3, randomized, parallel-group study comparing Sativex with placebo in people with multiple sclerosis and symptoms of spasticity. The registry identifies the design as quadruple masked. It lists 337 participants as enrolled, with 166 in the Sativex arm and 169 in the placebo arm contributing to the posted primary efficacy analysis. These analysis counts reflect the population specified for that endpoint: randomized participants who received at least one actuation of study medication and had on-treatment efficacy data.

The registry lists the study start as 2005-03 and primary completion as 2005-12, and its status is completed. The sponsor is Jazz Pharmaceuticals. The record describes Sativex as active treatment and placebo as control. It does not report the treatment dose, the precise allocation ratio, the method used to generate or conceal the randomization sequence, the number or location of study sites, or a detailed schedule of study visits.

The primary endpoint was change from baseline in mean spasticity Numerical Rating Scale score during the last 14 days of treatment, with the primary timeframe listed as 0-15 weeks. The sponsor’s reported comparison used ANCOVA and presented an estimated mean treatment difference. Separate posted arm summaries allow an unadjusted mean-difference calculation. Those are different statistical summaries: the ANCOVA estimate is the sponsor’s reported adjusted analysis, while the recalculation compares the posted group summaries without adjustment.

The registry lists eight secondary outcomes and nine outcome measures overall. Posted analyses are available for the primary endpoint and three secondary outcomes. The registry record does not report the original sample-size justification or a full protocol-level statistical analysis plan. It also does not report whether interim analyses were conducted, how missing endpoint values were handled beyond the stated primary analysis population, or whether adjustments were planned for testing multiple outcomes.

4. Participant Flow and Accounting

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

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: All subjects who were randomised, received at least one actuation of study medication and had on-treatment efficacy data were included in the analysis

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.

6. Primary Endpoint

The registry lists the primary outcome as Change From Baseline in Mean Spasticity 0-10 Numerical Rating Scale (NRS) Score During the Last 14 Day of Treatment (End of Treatment), measured over 0-15 weeks.

The primary endpoint is change from baseline in mean spasticity score on a 0-to-10 Numerical Rating Scale during the last 14 days of treatment. Participants were asked each evening to rate their spasticity over the preceding 24 hours, where 0 meant no spasticity and 10 meant the worst possible spasticity. The registry states that a negative value for change indicates improvement from baseline.

A daily rating captures a symptom from the participant’s perspective and can reflect how spasticity is experienced in ordinary life. Using the mean during the final 14 days focuses the endpoint on a sustained period near the end of treatment rather than on a single daily response. Comparing change from baseline between randomized groups is intended to account for the fact that participants may begin with different symptom scores. The endpoint is continuous: it retains the magnitude of score changes rather than reducing each person’s response to an improved/not-improved category.

The posted group means for change were -1.22 for Sativex and -0.91 for placebo. The sponsor’s reported ANCOVA estimate of the mean treatment difference was -0.23, with a 95% confidence interval from -0.59 to 0.14. The registry does not state a threshold for a clinically important change on this scale. Thus, the score definition explains the direction of change, but the registry alone does not establish how large a difference would matter to participants.

7. Results Posted in the Registry

• Sativex: mean -1.22 versus -0.91 on control; difference -0.31 (95% CI -0.68 to 0.06), standardized effect d = -0.18; z-test p = 0.103.

8. Secondary Efficacy Endpoints

The registry also posts results for secondary outcomes. For each outcome below, the endpoint type, the test applied to the posted summary statistics, the result and its interpretation are given in turn. Secondary endpoints are usually not adjusted for multiple testing, so they support rather than confirm the primary finding.

Number of Subjects With a 30% or Greater Improvement in Mean Spasticity 0-10 NRS Score at the End of Treatment Compared to Baseline

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Sativex: 30.7% (51/166) versus 24.9% (42/169) on control; risk difference 5.9 percentage points (95% CI -3.7 to 15.4), risk ratio 1.24 (0.87–1.75); unadjusted two-proportion z-test p = 0.230.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Sativex and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline in the Mean Modified Ashworth Scale Score at the End of Treatment

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Sativex: mean -3.30 versus -2.80 on control; difference -0.50 (95% CI -2.39 to 1.39), standardized effect d = -0.06; z-test p = 0.604.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Sativex and the control. The interval gives the range of effects that remain compatible with the data.

Change From Baseline in Mean Sleep Quality 0-10 NRS During the Last 14 Days of Treatment (End of Treatment)

• Endpoint type: Continuous: a mean value per participant, compared as a difference in means.
• Test: Unadjusted two-sample z-test for the difference in means, using the posted standard deviations.
• Result:
  • Sativex: mean -0.70 versus -0.60 on control; difference -0.10 (95% CI -0.74 to 0.54), standardized effect d = -0.04; z-test p = 0.759.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Sativex and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes address several related but distinct aspects of symptoms and function. Posted analyses include a threshold-based spasticity response measure and continuous changes in modified Ashworth Scale score and sleep-quality score. The registry also lists outcomes involving walking time, daily activities, carer impression, adverse events, and a higher spasticity-improvement threshold. These endpoints do not all measure the same underlying construct, and their results cannot be treated as interchangeable replications of the primary scale.

Across the posted secondary results, the estimates do not form a uniformly decisive pattern that changes the interpretation of the primary result. Some arm summaries numerically favor Sativex, but the comparisons have uncertainty and should be read using their own reported measures and intervals. The primary endpoint remains the central registered efficacy comparison; secondary measures can add context about whether related outcomes point in a consistent direction, but they do not remove uncertainty in the primary estimate.

The secondary results carry less inferential weight when several outcomes are assessed separately without a reported adjustment for multiplicity. Testing multiple outcomes increases the possibility that at least one apparently notable result occurs by chance, even if there is no true difference for any outcome. The registry record does not report a multiplicity-control plan, a hierarchy for secondary testing, or which secondary outcomes were designated as confirmatory. Consequently, individual secondary p values should not be treated as definitive evidence in isolation.

There are eight listed secondary outcomes, while posted analyses are available for three of them. The record does not explain why results for the other listed outcomes are absent. That gap limits assessment of the full pattern across the outcomes named in the registry. In general, secondary evidence is most informative when its measures, analysis plan, and handling of multiple tests are prespecified and fully reported; those details are not reported in the registry record.

Secondary outcomes registered for this trial, for reference (only those above were recalculated):

• Number of Subjects With a 30% or Greater Improvement in Mean Spasticity 0-10 NRS Score at the End of Treatment Compared to Baseline
• Change From Baseline in the Mean Modified Ashworth Scale Score at the End of Treatment
• Change From Baseline in Mean Sleep Quality 0-10 NRS During the Last 14 Days of Treatment (End of Treatment)
• Incidence of Adverse Events as a Measure of Subject Safety
• Change From Baseline in Mean Timed 10 Metre Walk Time at the End of Treatment
• Carer Global Impression of Change at the End of Treatment
• Change From Baseline in the Mean Total Barthel Activities of Daily Living Index Score at the End of Treatment
• Number of Subjects With a 50% or Greater Improvement in Mean Spasticity 0-10 NRS Score at the End of Treatment Compared to Baseline

9. Statistical Interpretation

10. Statistical Methods Explained

The primary measure is a continuous outcome: a numerical score that can take a range of values. For each group, the registry reports the number contributing data, the mean change from baseline, and the standard deviation. A mean describes the group’s average change. A standard deviation describes how spread out individual values are around that average; it is not the uncertainty of the group mean and is not a confidence interval. The posted primary means are -1.22 for Sativex and -0.91 for placebo, with standard deviations of 1.76 and 1.72, respectively.

What does the mean difference represent?

The unadjusted mean difference compares the two arm means, with the Sativex group treated as the intervention and placebo as the reference. The recalculated estimate is -0.31. Under the scale definition, a negative change from baseline indicates improvement, so a negative contrast points toward a greater average reduction in the Sativex group in this unadjusted comparison. That direction is a description of the estimate, not proof of a treatment effect. The uncertainty interval and the sponsor’s reported model result also matter.

The recalculation is unadjusted: it uses posted summary statistics rather than participant-level data or a regression model. It does not account for baseline covariates or other model features. The registry gives a pooled standard deviation of 1.74 for this recalculation and an unadjusted p value of 0.103. These values characterize the posted-summary comparison; they do not reproduce every feature of the sponsor’s analysis.

Why is ANCOVA reported separately?

The sponsor reports ANCOVA, or analysis of covariance, for the primary comparison. In general, ANCOVA compares group outcomes while adjusting for specified covariates, which can include a baseline measure when the analysis plan calls for it. Adjustment can change the treatment estimate and its precision relative to a simple comparison of group summaries. The sponsor’s estimated mean treatment difference is -0.23, with a 95% confidence interval from -0.59 to 0.14 and p = 0.220. The registry does not report the full model specification, covariates, or detailed handling of missing observations, so those aspects cannot be assessed from the posted record.

The ANCOVA result and the unadjusted recalculation answer closely related but not identical statistical questions. The former is the sponsor’s model-based estimate; the latter is a straightforward comparison of summary means. The registry record does not flag a discrepancy between the posted results, but that does not make the analyses interchangeable. When interpreting the trial, the sponsor-reported ANCOVA result remains the relevant reported analysis, while the recalculation offers a transparent comparison based on the available arm summaries.

How should the p value be understood?

A p value is calculated under a statistical model and a null hypothesis, usually that the groups do not differ on the measure being tested. It assesses how compatible the observed data are with that null model. It does not provide the probability that the null hypothesis is true, nor does it measure whether a difference matters to participants. The sponsor’s p value is 0.220; the unadjusted recalculated p value is 0.103. Both should be read together with their confidence intervals and the methods that generated them, not as stand-alone judgments.

The registry does not report the original trial’s full statistical analysis plan or a prespecified primary decision threshold. The planning illustration lists a two-sided significance level of 0.05, but that illustration is a separate calculation based on the observed effect and is not documentation of the trial’s original design assumptions.

What do the planning calculations describe?

The sample-size illustrations use the observed effect as a reference and show how the required number of participants per arm changes when planning for 100%, 75%, or 50% of that effect. The stated assumptions are two-sided significance level 0.05, equal allocation between the two arms, and a normal approximation. These assumptions belong to the planning illustration; they should not be attributed to the original trial. The planning values are hypothetical design calculations, not evidence about the probability that this completed trial detected an effect.

In general, power is the probability that a study design will detect an effect of a specified size if that effect is present and the model assumptions hold. The 80% and 90% columns describe different target probabilities for detecting the chosen effect under the planning assumptions. Smaller target effects generally require more participants to distinguish them from random variation. These calculations do not turn the observed result into a measure of the trial’s actual power, and they do not show that the trial was or was not adequately designed.

Why do endpoint type and model choice matter?

The primary endpoint is continuous, so the principal comparison concerns a mean difference. A secondary outcome may instead count participants who crossed a response threshold. For such a binary outcome, an absolute risk difference compares the proportions meeting the threshold, while an odds ratio compares odds. These measures are not interchangeable: an odds ratio is not a direct statement of how many additional participants responded, and the absolute difference depends on the underlying event proportions. The registry reports different analysis summaries for different outcomes, so interpretation should follow the measure and model reported for each one.

For all of these methods, statistical precision depends on the amount of information in the data, the variability of the outcome, and the analysis model. A confidence interval expresses a range of estimates compatible with the data and model at its stated confidence level. It conveys more about uncertainty than a p value alone, while still relying on the assumptions of the analysis. The registry does not provide participant-level data or enough model detail to independently check every assumption.

11. Confidence Intervals and Precision

The sponsor-reported primary estimate is -0.23, with a 95% confidence interval from -0.59 to 0.14. The interval crosses zero, the value corresponding to no mean difference between groups. It therefore includes estimates in either direction: values below zero favor a greater reduction in the Sativex group under the reported contrast, while values above zero point in the other direction. The interval does not establish that the true difference is zero, and it does not exclude every potentially meaningful effect.

A confidence interval describes the statistical uncertainty around an estimate under the analysis model. In repeated studies conducted under the same assumptions, intervals constructed in this way would cover the true parameter at the stated rate. For an individual study, it is best read as a range of values reasonably compatible with the observed data and the model, rather than as a probability distribution for the true effect. The breadth of the interval matters: estimates near its ends are less directly supported than the central estimate, but remain within the reported range.

The unadjusted recalculation gives a mean difference of -0.31 and a 95% confidence interval from -0.683 to 0.063. This interval also includes zero. Its endpoints differ from the sponsor’s ANCOVA interval because it is based on a separate, unadjusted calculation. The two intervals should not be combined or treated as repeated estimates from the same model.

Precision is not the same as clinical importance. The registry does not report a threshold for a meaningful change in the spasticity score, so it is not possible to classify the full interval as excluding or including a clinically important effect. The estimates and intervals describe statistical uncertainty in the reported contrast; they do not settle the clinical interpretation on their own.

12. Planning a Similar Trial: Sample Size

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

The planning figures are participants per arm for this continuous primary outcome, not total events. The planning assumptions stated for these illustrations are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set at the listed fraction of the observed effect. These are planning assumptions for the illustrations; the registry record does not report the original trial’s sample-size justification or its own design assumptions.

The 80% power column gives the estimated participants per arm for a design intended to have an 80% chance of detecting the specified effect under those assumptions, if that effect is truly present. The 90% power column uses a higher target chance of detection for the same specified effect. Both are design targets, not descriptions of what happened in the completed trial. For the primary comparison, the display gives 495 participants per arm for 80% power and 663 for 90% power when planning for 100% of the observed effect. For 75% of the observed effect, the figures are 880 and 1,177; for 50%, they are 1,979 and 2,649.

Smaller effects are harder to distinguish from random variation, so planning to detect a smaller fraction of the observed effect requires more participants. A curve expressing the same planning calculations should be read as a visual depiction of how estimated sample size changes with the target effect and power, under the stated assumptions. The calculations are not observed power and do not tell whether the completed trial was adequately designed. They are hypothetical planning illustrations based on the observed estimate, not the trial’s original sample-size plan.

13. How the Sample Size Is Derived

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

Formula

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

Worked calculation for this trial

1. Observed difference Δ = -0.310; pooled standard deviation σ = 1.740.
2. n = 2 × (1.96 + 0.842)² × 1.740² / -0.310² = 494.5, rounded up to 495 participants per arm for 80% power.

The smaller-effect rows in the table above repeat the same calculation with the effect multiplied by 0.75 and 0.5 (for ratios, applied to the logarithm of the ratio).

R code

alpha <- 0.05
z_a   <- qnorm(1 - alpha / 2)          # two-sided
delta <- -0.31                      # observed difference in means
sigma <- 1.73993476112554                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^2, 6))
}
for (f in c(1, 0.75, 0.5))
  cat(sprintf("%3.0f%% of observed effect: n per arm = %s (80%% power), %s (90%% power)\n", 100 * f,
              format(n_per_arm(0.80, f), big.mark = ","), format(n_per_arm(0.90, f), big.mark = ",")))

# Check against R's built-in function (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(1599234)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 495 (80% power), 663 (90% power)
 75% of observed effect: n per arm = 880 (80% power), 1,177 (90% power)
 50% of observed effect: n per arm = 1,979 (80% power), 2,649 (90% power)
power.t.test: 495.5 
Simulated power at n = 495 per arm: 0.800 (20000 replicates)

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

14. Analyses Reported by the Sponsor

The sponsor’s reported primary analysis is ANCOVA, with an estimated mean treatment difference of -0.23, a 95% confidence interval from -0.59 to 0.14, and p = 0.220. This is the sponsor’s adjusted, model-based comparison and should be treated as the primary reported inferential result. The registry does not describe all model covariates or the full analysis specification.

The unadjusted recalculation from the posted arm summaries estimates a mean difference of -0.31, with a 95% confidence interval from -0.683 to 0.063 and p = 0.103. It does not apply the ANCOVA model and therefore is not a reproduction of the sponsor’s adjusted analysis. Both estimates are negative and both confidence intervals include zero, but their numerical values and p values differ because they represent distinct analyses.

The registry record does not identify a discrepancy between the sponsor’s result and the posted summary data. The recalculation is useful as a transparent check based on the arm means and variability, while the sponsor’s ANCOVA result remains the analysis reported by the sponsor. Neither estimate alone establishes the clinical importance of the difference; the registry does not provide a clinical-importance threshold for this scale.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when a study evaluates several outcomes, comparisons, time points, or subgroups. If each result is interpreted as though it were the only test, the chance of obtaining at least one apparently unusual result can increase. Statistical approaches can control this issue, or an analysis plan can establish a hierarchy in which some outcomes are confirmatory and others exploratory. The registry record lists a primary endpoint and eight secondary outcomes, but it does not report whether the original analysis plan used a multiplicity adjustment or testing hierarchy.

The primary endpoint is therefore the clearest registered focus for interpreting the efficacy results. The secondary findings can describe whether related measurements point in a similar direction, but their individual p values should be treated cautiously without information about planned multiplicity control. The registry also does not state whether all analyses were prespecified, whether any additional analyses were conducted, or how the posted analyses were selected.

The record identifies the primary analysis population as randomized participants who received at least one actuation of study medication and had on-treatment efficacy data. That definition is more specific than an analysis of every randomized participant regardless of available efficacy measurements. The registry does not report how many randomized participants lacked the required data, reasons for missingness, the amount of missing data by group, or the method used to address missing values. It is therefore not possible to assess whether missing data could have affected the comparison.

The registry reports the sponsor’s ANCOVA method but not the full model specification, the covariates included, or any sensitivity analyses. It also does not report the original sample-size justification, interim analyses, or a complete statistical analysis plan. These omissions do not show that a particular procedure was or was not used; they mean the corresponding details are not reported in the registry record. Without them, readers cannot fully assess prespecification, robustness to missing observations, or the role of multiple testing in the interpretation of secondary results.

16. 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 3% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: urinary tract infection (1.8%), multiple sclerosis relapse (1.2%), erysipelas (1.2%), sepsis (1.2%), irritability (0.6%).

The registry reports serious adverse events affecting 15 of 167 participants at risk in the Sativex group and 7 of 170 at risk in the placebo group. These are counts of participants with a serious adverse event and the corresponding numbers at risk. They are safety summaries, not part of the primary efficacy comparison. The registry lists urinary tract infection at 1.8%, multiple sclerosis relapse at 1.2%, erysipelas at 1.2%, sepsis at 1.2%, irritability at 0.6%, and asthenia at 0.6% among the top serious adverse event terms.

The record also reports other adverse events affecting 156 of 167 participants at risk in the Sativex group and 132 of 170 at risk in the placebo group. The posted serious-event figures alone do not describe severity patterns beyond the serious-event classification, timing, duration, or the clinical context of each event. Death counts are not reported in the registry record; a missing death value should not be interpreted as confirmation that no deaths occurred.

These counts should be read as descriptive safety information from the registry. The record does not provide enough detail to determine whether group differences reflect treatment, underlying illness, exposure time, or other factors. Safety interpretation depends on the full event definitions, follow-up, and clinical review, not counts in isolation.

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

The primary result describes a difference in average change in participant-rated spasticity between the Sativex and placebo groups. The sponsor’s ANCOVA estimate is -0.23, and its 95% confidence interval extends from -0.59 to 0.14. The interval includes zero, so the result does not establish a clear difference between groups on this endpoint. The unadjusted recalculation is also uncertain, with a 95% confidence interval that includes zero.

This does not mean the groups are proven to be identical. A confidence interval that crosses zero can be compatible with no average difference and with differences in either direction within the interval. Nor does the result establish that any possible difference would be important to participants. The registry does not report a threshold for a clinically meaningful change in the 0-to-10 spasticity rating, so statistical estimates cannot be translated into a judgment about meaningful benefit from the record alone.

The arm averages moved in the negative direction, which the registry defines as improvement from baseline. That descriptive pattern does not by itself show that the intervention caused the change. The randomized placebo comparison is the relevant contrast, and its uncertainty needs to be considered alongside the group summaries. The sponsor’s reported adjusted analysis and the unadjusted recalculation are separate estimates, not competing versions of a single identical calculation.

The secondary outcomes provide information about related symptoms and function but do not resolve the uncertainty in the primary result. Multiple secondary analyses also warrant caution because the registry does not report how multiple testing was handled. The safety counts describe reported events but cannot by themselves establish a causal relationship or provide a complete clinical safety assessment.

In short, the registry supports a precise description of the measured changes and their statistical uncertainty, but not a definitive claim about clinical benefit, lack of benefit, or safety. The original analysis plan, clinical-importance threshold, complete missing-data information, and additional protocol details are not reported in the registry record. Those limits constrain conclusions about why the estimates arose and how broadly they should be interpreted.

18. Why This Trial Matters Statistically

This trial illustrates why an arm-level change and a between-group contrast answer different questions. Both group means changed in the negative direction, which the registry defines as improvement. The clinical comparison, however, is whether the average change differed between randomized groups. A change within each arm does not on its own identify a treatment effect; the contrast and its uncertainty are central.

It also shows why an adjusted sponsor analysis and an unadjusted recalculation should be distinguished. The ANCOVA estimate uses the sponsor’s reported model, while the recalculation compares posted summary statistics. Their estimates are not identical, and the registry does not provide enough model detail to explain the difference. Reporting both transparently helps readers see how the choice of analysis affects the numerical summary without treating the recalculation as a replacement for the reported analysis.

Confidence intervals are important because a point estimate alone can suggest more certainty than the data support. The sponsor’s interval spans values on either side of zero. That makes the uncertainty visible and guards against interpreting a single estimate as a settled answer. The absent clinical-importance threshold also highlights that statistical uncertainty and clinical interpretation are related but distinct questions.

The trial further illustrates the challenges of interpreting a collection of secondary outcomes. Different measures may address symptoms, function, and safety, but examining many results without a reported multiplicity plan can make isolated findings difficult to interpret. The number of listed outcomes and the incomplete set of posted secondary analyses make full assessment of the outcome pattern more difficult. These are statistical reporting considerations, not evidence for or against a particular treatment conclusion.

19. 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 trial-specific details needed for a fuller assessment are not reported in the registry record. The original sample-size justification is absent, so the enrollment of 337 cannot be evaluated against a documented target effect or prespecified precision goal. The planning illustrations based on the observed effect do not reconstruct the trial’s original design assumptions.

The primary analysis population required participants to have received at least one actuation and to have on-treatment efficacy data. The record does not provide the number excluded from the analysis for lack of data, reasons for missing observations, or methods used to address missingness. This limits assessment of whether the analyzed population could differ from all randomized participants in ways relevant to the outcome.

The sponsor reports ANCOVA, but the registry does not report the covariates, model specification, or sensitivity analyses. The original statistical analysis plan, treatment dose and schedule, allocation ratio, randomization implementation details, study sites, and participant characteristics at baseline are also not reported in the registry record. These omissions prevent a detailed review of model assumptions, balance, treatment exposure, and generalizability.

The registry lists eight secondary outcomes, but posted analyses are available for three. It does not explain the absence of results for the others or report a multiplicity strategy. A clinical-importance threshold for the primary scale is not reported either, making it difficult to assess whether estimates within the confidence interval would matter to participants. These limitations qualify interpretation of the posted findings without establishing that the analyses were inappropriate or that unreported procedures were not performed.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study to Evaluate the Efficacy of Sativex in Relieving Symptoms of Spasticity Due to Multiple Sclerosis — NCT01599234.
• Registry record last updated 2023-05-03; status last verified 2023-04.
• Results first posted 2012-08-16.
• 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.

23. Record Summary

The registry describes a completed, randomized, quadruple-masked, parallel-group phase 3 study of Sativex versus placebo for spasticity symptoms in multiple sclerosis. Enrollment was 337. The primary endpoint was change from baseline in mean self-rated spasticity score during the last 14 days of treatment, on a 0-to-10 scale.

The sponsor reported an ANCOVA treatment difference of -0.23, with a 95% confidence interval from -0.59 to 0.14 and p = 0.220. The unadjusted recalculation was -0.31, with a 95% confidence interval from -0.683 to 0.063 and p = 0.103. Both intervals include zero. The registry also posts analyses for three secondary outcomes, while listing eight secondary outcomes overall.

The registry does not report the original sample-size justification, full statistical analysis plan, multiplicity approach, or detailed handling of missing efficacy data. Those omissions limit interpretation beyond the reported estimates, intervals, and analysis population.