The MOTS (Medication Overuse Treatment Strategy) Trial: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The MOTS (Medication Overuse Treatment Strategy) Trial (NCT02764320), a randomized na study. The primary outcome is Frequency of Moderate to Severe Headache Days.

2. Clinical Question

The registry describes a randomized comparison of two treatment strategies for people with chronic migraine with medication overuse. The strategies both include initiating or optimizing migraine prophylactic therapy; the stated distinction is whether the overused medication is discontinued immediately or is not discontinued early. The registry does not report which specific prophylactic medicines or overused medicines were used, or what criteria guided treatment choices.

The primary outcome counts days on which a headache lasts at least 2 hours and reaches moderate or severe intensity at some point. The reported group means were 9.3 days for discontinuation and 9.1 days for preventive therapy only. The posted mean difference for preventive therapy only compared with discontinuation was -0.2 days. Its confidence interval includes values on either side of zero, so the estimate does not precisely distinguish the strategies in either direction.

This is a question about the average frequency of a defined headache-day outcome over the specified period. It does not, by itself, address every aspect of migraine symptoms, medication use, or individual response. Details about eligibility beyond the broad population description, outcome collection procedures, and the analysis population are not reported in the registry record.

3. Trial Design

The MOTS trial is registered as NCT02764320 and is listed as completed. Mayo Clinic is the sponsor. The record lists an enrollment of 720, a start date of 2017-01-09, and a primary completion date of 2021-03-31. The phase is listed as NA.

The design is randomized, parallel-group, and unmasked. Participants were assigned to one of two strategies: discontinuation, in which migraine prophylactic therapy was initiated or optimized with immediate discontinuation of overused medication, or preventive therapy only, in which prophylactic therapy was initiated or optimized without early discontinuation. The registry does not describe the randomization sequence, allocation concealment, or operational details of treatment delivery. It lists no masking.

The primary outcome is a continuous measure reported as a mean and standard deviation. The outcome period is 12 weeks. The posted results give a mean of 9.3 days with a standard deviation of 7.2 in the discontinuation group, and a mean of 9.1 days with a standard deviation of 6.8 in the preventive therapy only group. The outcome entries list 294 and 291 participants, respectively; the record notes that outcome data were not collected or analyzed for 16 people in the discontinuation arm and 19 in the preventive therapy only arm.

The registry lists one outcome measure and no secondary outcomes. It lists no analysis entries, although the results include group summaries and a comparison for the primary outcome. The original sample-size justification, planned statistical analysis, and details of the intended analysis population are not reported in the registry record.

4. Participant Flow and Accounting

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

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Data not collected or analyzed on 16 subjects from the discontinuation arm. Data not collected or analyzed on 19 subjects from the preventive therapy only arm.

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 Frequency of Moderate to Severe Headache Days, measured over 12 weeks.

The primary endpoint is the frequency of moderate to severe headache days during 12 weeks. A day qualifies when a headache lasts at least 2 hours and reaches moderate or severe intensity at any time. The outcome is reported as a mean number of days, with a standard deviation describing the spread of individual outcome values around each group mean.

Counting days captures how often the specified headache experience occurred over the outcome period. Because the trial compares two strategies for chronic migraine with medication overuse, this measure offers a direct way to describe headache frequency in each group. It is a continuous outcome in the registry’s statistical classification, and the reported comparison focuses on the difference between the group means.

The definition does not capture every feature of headache burden. It does not, on its own, describe duration beyond the threshold in the definition, peak intensity beyond the moderate-or-severe classification, or other symptoms and consequences. The registry description does not state how headache days were recorded, how missing daily information was handled, or whether outcome assessors were involved in collection. Those measurement details are not reported in the registry record.

7. Results Posted in the Registry

• Preventive Therapy Only: mean 9.10 versus 9.30 on control; difference -0.20 (95% CI -1.33 to 0.93), standardized effect d = -0.03; z-test p = 0.730.

8. Statistical Interpretation

9. Statistical Methods Explained

A mean is the arithmetic average of the observed outcome values in a group. For this trial, the outcome is a count of qualifying headache days over the stated 12-week period, and the registry reports the mean for each strategy. The mean difference expresses the contrast between those averages. The posted comparison uses preventive therapy only minus discontinuation, so a negative value corresponds to a lower mean in the preventive therapy only group. This sign convention matters: the numerical direction depends on which group is subtracted from which.

A standard deviation describes how dispersed participant outcomes are around their group mean. It is not the uncertainty in the mean itself, and it is not a confidence interval. The reported standard deviations, 7.2 days for discontinuation and 6.8 days for preventive therapy only, indicate that individual outcomes varied around their respective means. A standard deviation does not tell whether variation is balanced across the two sides of the mean, nor does it show the full distribution. The registry does not provide distribution plots or participant-level data.

A confidence interval gives a range of values that are reasonably compatible with the estimate under the method used to construct it. In repeated use of a correctly specified interval procedure, the stated confidence level has a long-run coverage interpretation. It is not a range that contains a known proportion of individual outcomes, and it is not a probability statement about the particular true value after the data have been observed. The registry does not state the confidence level or the specific method used for the posted interval, so those details should not be inferred.

The comparison estimate is -0.2 days, and its interval extends from -1.335 to 0.935 days. Because the interval crosses zero, it includes a value consistent with no mean difference as well as values in either direction. Its width is relevant to precision: the estimate does not pin down a narrow range of possible mean differences. An interval’s interpretation also depends on the study design, outcome measurement, analysis population, and assumptions behind the calculation. Several of those details are not reported in the registry record.

The posted p-value is unadjusted. In general, a p-value evaluates how unusual the observed data or a more extreme result would be under a specified null model, if the model assumptions hold. It does not measure effect size or clinical importance. It is also not the probability that the null hypothesis is true. A p-value should therefore be considered together with the estimated difference and its interval, rather than read as a standalone verdict. The registry does not report the planned test or its assumptions.

Random assignment is intended, in general, to make treatment groups comparable at the point of assignment by distributing both measured and unmeasured characteristics through chance. It does not guarantee identical groups, and it does not address bias introduced later through missing outcome data, adherence, or measurement. This trial is listed as randomized and unmasked. The record does not describe allocation concealment or how treatment knowledge may have affected outcome collection or analysis.

The parallel design means the two strategies are compared across separate groups rather than by having the same participants receive both strategies in sequence. The registry’s outcome summaries give arm-specific means and standard deviations, while the posted comparison gives a mean difference, interval, and unadjusted p-value. The record does not describe whether the comparison adjusted for baseline variables, how any departures from model assumptions were assessed, or what analysis population was intended. It also lists no analysis entries, so the posted summaries do not substitute for a full statistical analysis plan.

Over-dispersion is a general concern when count outcomes vary more than a simple count model assumes. In general, a model that understates variation can produce overly narrow uncertainty intervals or p-values that are too small. The posted endpoint is summarized as a mean with a standard deviation, but the registry does not report a count-model analysis, a distributional assessment, or an over-dispersion adjustment. It is therefore not possible to say from the record whether those methods were considered.

The registry identifies the primary outcome as continuous and reports its mean and standard deviation. That is a descriptive summary choice; it does not establish which inferential method was used. The sample-size illustrations use a normal approximation, as stated in the registry record, but that planning assumption should not be mistaken for confirmation that the primary outcome comparison itself used a particular normal-based test. The trial’s original analysis and design assumptions are not reported.

10. Confidence Intervals and Precision

The estimated mean difference is -0.2 days for preventive therapy only compared with discontinuation. Its confidence interval extends from -1.335 to 0.935 days. The interval includes zero and values on both sides of it, so the estimate is not precise enough to establish a clear direction of difference from the posted comparison alone.

The range also matters for judging what the result can rule out. It allows for a lower mean in either group, rather than restricting plausible values to a narrow range around the point estimate. The group standard deviations, 7.2 and 6.8 days, are much larger than the absolute point estimate, underscoring that the group means sit within considerable outcome variation. This comparison of scales is descriptive; it does not replace the interval or establish the importance of any particular difference.

The registry does not state the confidence level or construction method for the interval. It also does not report the prespecified analysis population, missing-data method, or whether the posted comparison was adjusted. Those omissions limit a full assessment of the interval’s assumptions. The unadjusted p-value is 0.73, but it should not be treated as a measure of precision by itself. The interval gives a more direct view of the range of mean differences compatible with the posted analysis, while still depending on the method and data completeness.

11. 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 sample-size figures are planning illustrations, not a report of the trial’s original design calculation. They give participants per arm, not total participants or total events. The listed assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect used for planning is expressed as a stated fraction of the observed effect.

For planning at 100% of the observed effect, the display gives 19,252 participants per arm for 80% power and 25,772 participants per arm for 90% power. For planning at 75% of the observed effect, it gives 34,225 participants per arm for 80% power and 45,817 participants per arm for 90% power. For planning at 50% of the observed effect, it gives 77,005 participants per arm for 80% power and 103,088 participants per arm for 90% power.

In general, power is the chance that a planned statistical test will detect an effect of the specified size under the assumptions used, if that effect is truly present. The 80% and 90% columns describe different target chances of detection for a future study; the higher target generally requires more participants under otherwise matching assumptions. The listed scenarios also illustrate that smaller effects need more participants to detect reliably. The figures are not observed power.

The registry states that the original design assumptions are not represented by these calculations. The trial’s own sample-size justification and planning assumptions are not reported in the registry record. The illustrations are tied to the observed comparison and should not be treated as a recommendation or as evidence about the trial’s ability to detect a different effect under a different design.

12. 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 = -0.200; pooled standard deviation σ = 7.004.
2. n = 2 × (1.96 + 0.842)² × 7.004² / -0.200² = 19,251.1, rounded up to 19,252 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 <- -0.200000000000001                      # observed difference in means
sigma <- 7.00388522416495                     # 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(2764320)
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 = 19,252 (80% power), 25,772 (90% power)
 75% of observed effect: n per arm = 34,225 (80% power), 45,817 (90% power)
 50% of observed effect: n per arm = 77,005 (80% power), 103,088 (90% power)
power.t.test: 19252.1 
Simulated power at n = 19,252 per arm: 0.803 (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.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists one outcome measure, the primary frequency of moderate to severe headache days, and no secondary outcomes. It also lists no analysis entries. These registry entries do not establish that no other outcomes or analyses were planned; a multiplicity strategy, including how multiple outcomes or repeated analyses were to be handled, is not reported in the registry record.

The primary outcome description states that data were not collected or analyzed for 16 people in the discontinuation arm and 19 in the preventive therapy only arm. The arm summaries list 294 and 291 participants, respectively. The record does not explain why outcome data were absent, whether the absence arose before or after randomization, or whether any missing values were imputed. It also does not describe sensitivity analyses or methods for assessing the effect of missing data.

The analysis population is not identified as intention-to-treat, per-protocol, or another defined set. The record does not state whether participants were analyzed according to their randomized assignment, whether treatment adherence was considered, or how deviations from assigned strategy were handled. These distinctions can matter when interpreting a randomized comparison.

Safety reporting has its own at-risk counts, which differ from the outcome counts. The registry does not explain the relationship between these reporting sets or provide a general accounting of participant flow. Therefore, the outcome and safety denominators should be read as reported for their respective results, without assuming that they represent the same analysis population.

14. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 1% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: fracture (0.6%), anemia (0.3%), anxiety (0.3%), dyspnea (0.3%), fatigue (0.3%).

For serious adverse events, the registry lists 7 affected among 361 at risk in the discontinuation group and 7 affected among 359 at risk in the preventive therapy only group. For other adverse events, it lists 20 affected among 361 at risk and 16 among 359 at risk, respectively. These are descriptive counts from the safety results; they do not by themselves establish whether an event was caused by a treatment strategy.

The record lists zero deaths in the discontinuation group and 1 in the preventive therapy only group. It also names fracture at 0.6%, and anemia, anxiety, dyspnea, fatigue, and joint range of motion decrease cervical spine at 0.3% each among the top serious-event terms. The record does not identify the arm attribution for these listed terms, so they should not be assigned to a particular group.

Safety denominators are distinct from the primary-outcome counts, and the registry does not explain the difference. Event definitions, severity grading details, timing, and causal assessments are not reported in the registry record. The counts should therefore be read as registry safety summaries, not as a complete account of comparative risk.

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

The posted results describe a small difference between the group means for the primary outcome: 9.3 moderate to severe headache days in the discontinuation group and 9.1 in the preventive therapy only group over 12 weeks. The estimated mean difference is -0.2 days when preventive therapy only is compared with discontinuation. The confidence interval includes zero and extends in both directions, so the result does not precisely establish which strategy had the lower mean.

This is not evidence that the strategies are equivalent. Demonstrating equivalence or non-inferiority requires a suitable design, a prespecified margin, and an analysis aligned to that objective; those elements are not reported in the registry record. Nor does the posted comparison prove that there is no difference. It describes the estimate and uncertainty in the registry result.

The result is specific to the defined outcome and period. It does not establish how the strategies compare on longer-term headache frequency, other migraine symptoms, medication use, quality of life, or outcomes not posted in the registry. It also does not describe individual benefit or harm. Group averages can conceal variation among participants, and no subgroup or individual-level results are reported.

Missing primary-outcome data are noted for both arms, but their reasons and handling are not reported. The record also does not give the analysis population or full statistical plan. These limitations make it important to treat the posted figures as a registry summary rather than a complete account of the trial’s conduct and analysis. No treatment recommendation follows from these group summaries alone.

16. Why This Trial Matters Statistically

This trial poses a statistical comparison between two strategies that share a preventive-therapy component and differ in the timing of discontinuing overused medication. That makes the contrast interpretable as a comparison of the strategy packages described in the registry, rather than a comparison of prophylactic therapy against no prophylactic therapy. The record does not provide enough detail to separate the effects of individual medicines or other treatment decisions.

The primary outcome is a mean count of defined headache days, which makes the group contrast straightforward to describe. But a mean alone does not capture the range of individual experiences. The standard deviations are large relative to the mean difference, and the confidence interval includes values in both directions. Together, these statistics illustrate why the point estimate alone is not enough to assess certainty.

The result also highlights the distinction between failing to identify a clear difference and establishing equivalence. A non-significant-looking result or a confidence interval crossing zero does not show that two strategies have the same effect. A formal claim of equivalence requires a specified margin and suitable analysis; those are not reported in the registry record.

Finally, the missing outcome data and absent analysis details matter for interpretation. Randomization is a design strength in general, but its benefits do not remove the need to understand who contributed outcome data, how missingness was handled, and what analysis was planned. The record supports a careful reading of the posted estimate and its uncertainty, while leaving these methodological details unresolved.

17. 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 registry’s primary-outcome note says data were not collected or analyzed for 16 people in the discontinuation arm and 19 in the preventive therapy only arm. It does not give reasons, timing, or methods for handling those missing outcomes. Without that information, the possible influence of missing data on the mean comparison cannot be assessed from the record.

The registry describes the trial as unmasked and does not report whether headache-day information was collected by participant report, diary, or another method. It also does not describe the randomization process, allocation concealment, treatment adherence, crossovers, or departures from assigned strategy. These details would help evaluate implementation and potential sources of bias but are not reported in the registry record.

The outcome summaries and comparison are posted, but no analysis entries are listed. The planned statistical method, analysis population, interval construction, multiplicity approach, and any prespecified handling of model assumptions are not reported. The confidence interval and p-value can therefore be described as posted without claiming to know the full analysis procedure.

The registry provides one primary outcome and no secondary outcomes, leaving other possible measures of benefit and harm unavailable in the results record. It reports safety counts but does not explain how the safety population relates to the outcome population, and it does not provide arm attribution for the listed serious-event terms. The results also cover the stated 12-week outcome period; longer-term results are not reported in the registry record.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: The MOTS (Medication Overuse Treatment Strategy) Trial — NCT02764320.
• Registry record last updated 2021-11-12; status last verified 2021-11.
• Results first posted 2021-11-12.
• 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.

21. Record Summary

NCT02764320, the MOTS trial, is listed as completed and sponsored by Mayo Clinic. It was randomized, parallel-group, and unmasked, with an enrollment of 720. The registry lists a start date of 2017-01-09 and a primary completion date of 2021-03-31.

The trial compared immediate discontinuation of overused medication alongside initiated or optimized migraine prophylactic therapy with prophylactic therapy without early discontinuation. The primary outcome was mean frequency of moderate to severe headache days over 12 weeks. The reported means were 9.3 and 9.1 days, with a posted mean difference of -0.2 days and a confidence interval spanning -1.335 to 0.935 days.

The registry lists one outcome measure, no secondary outcomes, and no analysis entries. It notes missing outcome data in both arms, while reasons and handling are not reported. The posted comparison is therefore best read as a summary of the defined outcome and period, with uncertainty and reporting limitations kept in view.