A Study Comparing Oral Calcitonin to Nasal Spray Calcitonin in Postmenopausal Osteoporotic Women: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for A Study Comparing Oral Calcitonin to Nasal Spray Calcitonin in Postmenopausal Osteoporotic Women (NCT00959764), a randomized phase 3 study. The primary outcome is Percent Change From Baseline in Bone Mineral Density (BMD) of Axial Lumbar Spine.

2. Clinical Question

The registry record describes a comparison of effectiveness and tolerability among the oral tablet, nasal spray, and placebo regimens. The primary endpoint focused on bone mineral density rather than symptoms or fractures. The posted results report group-level means and standard deviations for this endpoint, alongside a sponsor-reported ANCOVA analysis across the three groups.

The posted primary-outcome summaries show a positive mean percent change in all three groups. The group means were 1.53 for oral calcitonin, 0.76 for nasal calcitonin, and 0.47 for placebo. These are descriptive summaries; they do not, on their own, establish a treatment effect or explain how the groups compare after adjustment. The sponsor-reported analysis and the unadjusted recalculation address group comparisons in different ways, so the distinction between them matters when interpreting the results.

The registry record does not report a non-inferiority margin or explain the precise clinical threshold that would make a difference in bone mineral density meaningful. It also does not report a sample-size justification. Those omissions limit conclusions about the trial’s original decision criteria, even though the endpoint and the posted statistical results are described.

3. Trial Design

This was a randomized, phase 3, parallel-group study with a placebo control. The official title describes it as double-blind, while the registry masking entry is quadruple. The registry record does not explain how masking was implemented or who was included in that masking designation. Enrollment is listed as 565. The study was completed, and its planned study duration was 48 weeks.

Three regimens were compared. One group received oral calcitonin tablets with placebo nasal spray. Another received active nasal calcitonin with an oral placebo tablet. The placebo group received matching placebo tablets and matching placebo nasal spray. The registry describes nasal calcitonin as an active comparator and notes in its summary that the nasal spray was available for osteoporosis treatment in the United States. The tested oral tablet contained salmon calcitonin.

The primary endpoint was percent change from baseline in axial lumbar spine bone mineral density. Bone mineral density was measured by dual-energy X-ray absorptiometry scans. Two scans were taken at each measurement timepoint and their mean was entered. The primary outcome timepoint was 48 weeks; if a participant did not complete the full study, the week 24 measurement was used as last observation carried forward.

The primary-outcome population was not simply everyone enrolled: the registry specifies randomization, receipt of treatment, and a post-baseline measurement taken at least 154 days after randomization. The posted primary group summaries include 189 participants in the oral group, 140 in the nasal group, and 82 in the placebo group. These endpoint counts should be read in the context of the stated analysis population, not as a replacement for the overall enrollment figure. The registry record does not report the original sample-size rationale, allocation ratio, or detailed sequence-generation method.

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: Unequal randomization was used for the study, with patients being assigned after inclusion/exclusion criteria were met.

Overall Study

Analysis population for the primary outcome, as defined in the registry: Patients who were randomized, received treatment, and had at least one post-baseline BMD value measured at least 154 days after randomization.

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 Percent Change From Baseline in Bone Mineral Density (BMD) of Axial Lumbar Spine, measured over 48 weeks.

The primary endpoint was percent change from baseline in bone mineral density of the axial lumbar spine, assessed at 48 weeks. Bone mineral density is a measure of bone mass per area, and dual-energy X-ray absorptiometry was used to obtain it. The registry set the baseline value to 0% for the change calculation. Two scans were taken at each timepoint, and the mean of those scans was entered for the measurement.

This endpoint was used to compare the groups’ change in a quantitative measure of bone density over the study period. It is a continuous outcome: each participant can have a value along a measurement scale, rather than being classified only as having or not having an event. That makes group means and measures of spread useful summaries, although a mean alone does not describe how individual responses varied.

The registry states that week 48 was the primary timepoint. If a participant did not complete the full study, the week 24 bone-density value was carried forward as the last observation. This is a missing-data handling rule, not an observed week 48 measurement; the resulting endpoint may therefore include values originating at different measurement times. The primary population also required a post-baseline measurement at least 154 days after randomization. The registry record does not report the number of participants whose endpoint value came from last observation carried forward, or the reasons for missing measurements.

The outcome is a surrogate measurement and is not itself a record of fractures, symptoms, or quality of life. Those outcomes are not reported as primary outcomes in this record.

7. Results Posted in the Registry

• Oral Calcitonin: mean 1.53 versus 0.47 on control; difference 1.06 (95% CI 0.23 to 1.89), standardized effect d = 0.33; z-test p = 0.012.
• Nasal Calcitonin: mean 0.76 versus 0.47 on control; difference 0.29 (95% CI -0.56 to 1.14), standardized effect d = 0.10; z-test p = 0.501.

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.

Change in Plasma C-terminal Telopeptide of Collagen 1 (CTx-1)

• 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:
  • Oral Calcitonin: mean -42.93 versus -21.09 on control; difference -21.84 (95% CI -36.97 to -6.71), standardized effect d = -0.45; z-test p = 0.005.
  • Nasal Calcitonin: mean -24.64 versus -21.09 on control; difference -3.55 (95% CI -19.09 to 11.99), standardized effect d = -0.07; z-test p = 0.654.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Oral Calcitonin than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Nasal Calcitonin and the control. The interval gives the range of effects that remain compatible with the data.

Change in Plasma CTx-1 From Baseline

• 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:
  • Oral Calcitonin: mean -29.92 versus -11.83 on control; difference -18.09 (95% CI -28.40 to -7.78), standardized effect d = -0.41; z-test p = <0.001.
  • Nasal Calcitonin: mean -11.41 versus -11.83 on control; difference 0.42 (95% CI -9.80 to 10.64), standardized effect d = 0.01; z-test p = 0.936.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Oral Calcitonin than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Nasal Calcitonin and the control. The interval gives the range of effects that remain compatible with the data.

Across the two posted secondary CTx-1 endpoints, the overall pattern is that the oral group has a more negative average percent change than placebo, while the nasal group’s averages are closer to placebo. The recalculated comparisons reflect that pattern: the oral-versus-placebo comparisons show estimated differences in the negative direction, while the nasal-versus-placebo comparisons are close to zero or have intervals spanning zero. These biomarker outcomes are distinct from the primary bone mineral density endpoint, so they provide related but not identical information.

The secondary results are broadly consistent with the primary result in the limited sense that the oral group’s reported outcomes tend to be more separated from placebo than the nasal group’s. They do not validate the primary analysis or establish a clinical consequence. CTx-1 is a plasma marker, and the posted outcomes are changes from baseline at 24 weeks and 48 weeks; they are not direct measures of fractures or patient experience.

Unadjusted secondary comparisons carry less weight when several outcomes and contrasts are examined without a reported multiplicity procedure. Testing multiple hypotheses increases the chance that some results will appear notable by chance. The registry record does not report whether multiplicity was addressed in the sponsor’s analysis, whether the endpoints were ordered in a testing hierarchy, or whether these secondary results were confirmatory or exploratory. The posted patterns should therefore be treated as supportive descriptive evidence, not as independent confirmation of the primary result.

The registry reports sponsor ANCOVA analyses for the secondary outcomes as well, with reported estimates and p-values. The record does not supply confidence intervals for those reported secondary analyses. Recalculated intervals from the summary statistics are not a replacement for model-based uncertainty estimates, and the p-values from sponsor and recalculated analyses are not directly interchangeable.

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

• Change in Plasma C-terminal Telopeptide of Collagen 1 (CTx-1)
• Change in Plasma CTx-1 From Baseline

9. Statistical Interpretation

10. Statistical Methods Explained

What does a group mean describe?

General explanation: a mean is the average of the observed outcome values in a group. For a continuous endpoint such as percent change in bone mineral density, it summarizes the group’s central tendency. It can be informative, but it does not show whether most participants had values close to the mean or whether responses were widely spread. The standard deviation is a measure of that spread: a larger standard deviation indicates more variation around the mean, while a smaller one indicates less variation. The posted primary group summaries report both means and standard deviations.

The primary outcome is reported as a least-squares mean. General explanation: least-squares means are model-based group estimates, often adjusted for variables included in a statistical model. The registry record does not list the covariates used to produce the primary least-squares means or provide enough detail to reconstruct the model. The displayed group summaries should therefore be distinguished from the sponsor’s named ANCOVA analysis and from the unadjusted recalculation.

What does ANCOVA contribute?

General explanation: analysis of covariance, or ANCOVA, compares group outcomes while accounting for specified covariates. When suitable covariates are included, adjustment can improve precision or account for baseline differences. The registry reports an ANCOVA result for the three primary groups, labelled as a net mean difference. The specific covariates, model diagnostics, and contrast definition are not reported in the registry record. It is not possible to infer those details from the label alone.

The sponsor’s reported primary estimate is 0.77, with a 95% confidence interval from 0.08 to 1.45 and p = 0.002. The registry marks this analysis as non-inferiority or equivalence. Such a classification depends on a prespecified margin and a defined comparison. The margin and further interpretive details are not reported in the registry record, so the label should not be expanded into a claim about satisfying an unstated criterion.

What is the recalculated comparison?

General explanation: an unadjusted comparison based on group summaries contrasts observed group means without accounting for covariates in a model. Here, the recalculated mean differences use the posted mean and standard deviation summaries and an approximate pooled standard deviation. These are simple unadjusted recalculations, not the trial’s reported ANCOVA estimates. For oral calcitonin versus placebo on the primary endpoint, the recalculated mean difference is 1.06, with p = 0.012. For nasal calcitonin versus placebo, the recalculated mean difference is 0.29, with p = 0.501.

The recalculation is not a reconstruction of the original analysis. It may differ from the sponsor result because the approach is unadjusted and uses summary statistics rather than participant-level data and the original model. The discrepancy flag is true; accordingly, the sponsor’s own analysis takes precedence for reporting the sponsor result. The recalculation remains a descriptive cross-check, and its assumptions should be kept visible when interpreting it.

How should p-values be understood?

General explanation: a p-value describes how compatible the observed data are with a specified null model, under the model’s assumptions. It is not the probability that the null hypothesis is true, and it does not measure the size or clinical importance of an effect. A smaller p-value can indicate that the observed data would be less typical under the null model, but it does not remove uncertainty about the estimated difference.

The primary sponsor analysis reports p = 0.002. The recalculated oral-versus-placebo comparison reports an unadjusted p-value of 0.012, whereas the recalculated nasal-versus-placebo comparison reports p = 0.501. These values refer to different analyses and should not be treated as interchangeable. The registry record does not report a multiplicity adjustment for the endpoint comparisons, nor does it state the complete testing hierarchy.

Why do the outcome and analysis population matter?

General explanation: the population used in an analysis defines whose measurements contribute to the estimate. The primary analysis population required randomization, treatment, and a qualifying post-baseline BMD measurement. That differs conceptually from a simple intention-to-treat population containing every randomized participant, and the record does not describe an intention-to-treat analysis for this endpoint.

The week 24 last-observation-carried-forward rule was used for participants who did not complete the full study. General explanation: carrying a prior observation forward fills a missing later measurement with an earlier recorded value. This can simplify analysis, but it treats that earlier value as the later outcome and may not reflect what would have been measured at the later time. Its implications depend on why values are missing and how outcomes change over time. The registry record does not report the missingness pattern, reasons for missing measurements, or sensitivity analyses using other approaches.

The posted analyses therefore combine a defined analysis population, a specified endpoint, and an incompletely described model. These features should be considered together. The means and recalculated comparisons are not a substitute for the sponsor’s analysis, and neither the model label nor the endpoint definition supplies unreported details about covariate selection, diagnostics, or robustness checks.

11. Confidence Intervals and Precision

A confidence interval gives a range of values compatible with an estimate under the statistical model and its assumptions. It communicates precision as well as direction: a narrower interval generally indicates a more precise estimate, while a wider interval leaves more uncertainty about the size of the difference. It is not a range containing every participant’s outcome, and it does not guarantee that the true value lies within the interval.

For the sponsor-reported primary ANCOVA result, the estimate is 0.77 and the 95% confidence interval extends from 0.08 to 1.45. This interval lies above zero, but its interpretation remains tied to the reported model and the “Mean Difference (Net)” contrast. The registry record does not define the contrast further or state the non-inferiority or equivalence margin. The p-value and interval are related summaries of this model-based result; neither supplies those missing design details.

The unadjusted recalculated comparison of oral calcitonin with placebo has an estimate of 1.06 and an interval whose endpoints are both positive. The nasal calcitonin comparison has an estimate of 0.29 and an interval that spans zero. These are recalculated intervals based on summary statistics, not the sponsor’s ANCOVA intervals. The underlying calculations use an approximate pooled standard deviation, so their precision should not be mistaken for precision from a fully adjusted participant-level model.

Group-level standard deviations show variability in observed outcomes, not the uncertainty in a treatment contrast. Also, the endpoint’s last-observation-carried-forward rule means some values may have come from week 24 rather than week 48. The registry record does not report how many endpoint values were carried forward or sensitivity analyses for that choice. Thus, statistical precision in the reported interval does not address every source of uncertainty in the measurement and analysis process.

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 estimates are expressed as participants per arm, not as total events. They are illustrations based on the observed effect, rather than the trial’s original design calculation. The stated assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. Each planning scenario asks how many participants per arm would be needed to detect a stated fraction of the observed effect under those assumptions.

The 80% power column gives the estimated number of participants per arm for a design that would have an 80% chance of detecting the assumed effect if that effect were real and the planning assumptions held. The 90% power column gives the corresponding planning target for a 90% chance. These are design probabilities under assumptions, not guarantees for a future study. The planning estimates are not observed power.

Rows that plan for 100% of the observed effect use the observed estimate as the target effect. Rows for 75% and 50% use smaller target effects. Smaller effects are harder to distinguish from random variation, so they require more participants per arm under the same assumptions. The estimates can therefore grow substantially as the target effect is reduced, particularly when the observed estimate is small.

The values are comparisons of two arms at a time, while the trial itself had three groups and the sponsor reported a three-group ANCOVA. The planning illustration is not a re-creation of that design or its analysis. The registry record does not report the original sample-size justification, the original target effect, or the trial’s design assumptions, so the planning figures should not be read as evidence about why the enrolled number was chosen or whether the trial met an original target.

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 = 1.060; pooled standard deviation σ = 3.182.
2. n = 2 × (1.96 + 0.842)² × 3.182² / 1.060² = 141.5, rounded up to 142 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 <- 1.06                      # observed difference in means
sigma <- 3.18209751730921                     # 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(959764)
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 = 142 (80% power), 190 (90% power)
 75% of observed effect: n per arm = 252 (80% power), 337 (90% power)
 50% of observed effect: n per arm = 566 (80% power), 758 (90% power)
power.t.test: 142.4 
Simulated power at n = 142 per arm: 0.794 (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-reported primary analysis is an ANCOVA across the three groups. It reports a “Mean Difference (Net)” of 0.77, a 95% confidence interval from 0.08 to 1.45, and p = 0.002. The registry classifies the analysis as non-inferiority or equivalence. It does not report the margin or enough detail to identify precisely how the net contrast was defined.

The recalculated primary comparison of oral calcitonin with placebo uses the posted group summaries, without adjustment, and gives a mean difference of 1.06 with an unadjusted p-value of 0.012. The recalculation is methodologically distinct from ANCOVA and is not an alternative estimate from the sponsor’s model. Differences between the two results can arise because an adjusted model and an unadjusted summary-statistic comparison do not estimate the contrast in the same way.

The discrepancy flag is true. The sponsor’s own analysis therefore takes precedence when describing the sponsor-reported result. The recalculation can still help readers see what follows from a simple comparison of the posted summaries, but it should not be presented as correcting or replacing the sponsor’s analysis. The record does not give the model details needed to independently reproduce the ANCOVA estimate.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when analyses examine multiple endpoints, groups, or contrasts. General explanation: the more hypotheses tested, the greater the chance that at least one result may appear unusual by chance if no adjustment or testing strategy is used. This does not mean every result is false; it means the interpretation should account for the number and relationship of the analyses.

The registry record includes a primary endpoint and two secondary CTx-1 endpoints. It reports analyses for the three-group study and also permits unadjusted recalculated comparisons against placebo. However, the record does not report whether the sponsor adjusted for multiple comparisons, defined a hierarchy of endpoints, or specified which analyses were confirmatory. The unadjusted secondary p-values should not be read as though they were protected from multiplicity by a prespecified procedure.

The record identifies the primary endpoint population as randomized participants who received treatment and had at least one post-baseline BMD measurement taken at least 154 days after randomization. It also states that week 24 BMD was carried forward for participants who did not complete the full study. It does not report the number of participants with missing primary endpoint measurements, the number whose values were carried forward, reasons for missingness, or sensitivity analyses. The secondary endpoint analysis populations are not reported in the registry record.

The record does not report the full statistical analysis plan, model covariates, diagnostic checks, interim analyses, or the original sample-size justification. It also does not report the non-inferiority or equivalence margin associated with the sponsor’s analysis classification. These omissions leave important parts of the analysis strategy unclear. They do not erase the posted estimates, but they constrain how confidently the results can be connected to prespecified decision criteria or interpreted as a complete account of all analyses.

Missing-data methods and multiplicity are separate concerns. Carrying a measurement forward affects how incomplete outcomes are represented; multiplicity concerns the interpretation of testing across outcomes and contrasts. Neither issue can be resolved from the reported p-values alone.

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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: cataract (1.0%), ileus (1.0%), biliary colic (1.0%), cholesystitis (1.0%), wrist fracture (1.0%).

The registry reports serious adverse event counts alongside the number at risk in each safety group. In the oral calcitonin group, 20 participants with 263 at risk were reported as affected; in the nasal calcitonin group, 9 with 182 at risk; and in the placebo group, 9 with 104 at risk. For other reported adverse events, the corresponding figures were 44 with 263 at risk, 146 with 182 at risk, and 83 with 104 at risk. These are counts, not calculated rates, and the categories should not be combined.

The listed top serious adverse event terms each have a reported frequency of 1.0%: cataract, ileus, biliary colic, Cholesystitis, wrist fracture, and spinal osteoarthritis. The record does not identify a treatment group for each listed term in this summary, so group-specific attribution should not be inferred.

Deaths are recorded as null for each group. That means death counts are not reported in the registry record; it does not mean that no deaths occurred. The registry summary does not provide event timing, severity beyond the serious-event classification, or individual causal assessments. These figures describe reported safety counts and should not be used alone to infer comparative risk.

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

The results describe changes in bone mineral density and plasma CTx-1 in a randomized comparison of oral calcitonin, nasal calcitonin, and placebo. On the primary endpoint, the posted group means are positive in all three groups, and the sponsor-reported ANCOVA gives a net estimate of 0.77 with a 95% confidence interval from 0.08 to 1.45. A separate unadjusted recalculation finds an oral-versus-placebo mean difference of 1.06. Because the analyses differ and the discrepancy flag is true, the sponsor analysis takes precedence as the sponsor’s result.

These measurements do not by themselves establish the effect on fractures, symptoms, quality of life, or other outcomes not reported as primary endpoints. The registry record does not provide a clinical threshold for interpreting the size of the BMD difference, or the margin underlying the non-inferiority or equivalence classification. Statistical evidence about a measured endpoint is not the same as evidence about every outcome a patient might value.

The findings also do not show that the groups are equivalent simply because a comparison has a nonsignificant p-value. A nonsignificant result can reflect uncertainty, and equivalence requires a defined margin and appropriate analysis. The margin is not reported. Nor can the unadjusted recalculations settle the question, because they do not reproduce the sponsor’s adjusted model.

Finally, the results need to be read alongside the analysis population and missing-data rule. Some primary values may have been carried forward from week 24, and the registry does not report how often this occurred. The posted statistics support a description of the measured and analyzed outcomes, with uncertainty, but do not justify claims beyond those outcomes or provide medical advice.

18. Why This Trial Matters Statistically

This trial illustrates why it is important to distinguish an endpoint summary from a model-based treatment comparison. The group means describe observed changes, while ANCOVA gives a model-based comparison whose interpretation depends on its contrast and covariates. A recalculation from summary statistics may be useful, but it is not interchangeable with the sponsor’s analysis. In this record, the discrepancy flag makes that distinction especially important.

It also shows how an analysis label can be informative but incomplete. The sponsor analysis is classified as non-inferiority or equivalence, yet the registry record does not report the margin that would define the decision. Without that element, readers cannot connect the label to a fully specified decision rule. Reporting the estimate, interval, and p-value is valuable, but the context for a non-inferiority or equivalence conclusion is also necessary.

The endpoint and missing-data procedure matter statistically. Bone mineral density was measured at scheduled timepoints, but the primary outcome could use a week 24 value carried forward when the full study was not completed. That choice can affect the interpretation of a nominal week 48 endpoint. The record does not report the amount of carry-forward or robustness analyses, which shows why transparent accounting of missing outcomes is important.

Finally, the secondary biomarker outcomes and multiple possible contrasts raise questions about multiplicity. Their patterns can add context, but without a reported testing strategy they carry less confirmatory weight. This record therefore offers a useful example of reading means, uncertainty, model labels, missing-data rules, and secondary analyses together rather than treating any single p-value as a complete 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.

The registry record leaves several trial-specific details unreported. It does not provide the original sample-size justification, allocation ratio, randomization sequence method, or full statistical analysis plan. The sponsor ANCOVA covariates, diagnostic checks, and precise definition of the “Mean Difference (Net)” contrast are also not reported. Those omissions limit independent reconstruction of the primary analysis.

The non-inferiority or equivalence designation lacks a reported margin. The record therefore does not establish the criterion against which that analysis was evaluated. Likewise, no prespecified clinically important difference for the BMD endpoint is reported. The primary endpoint’s interpretation is consequently limited to the measured changes and their statistical uncertainty.

Missing-data information is incomplete. The record specifies last observation carried forward for participants not completing the full study, but does not report how many values were carried forward, why participants did not complete follow-up, or sensitivity analyses using alternative assumptions. The primary analysis population has explicit eligibility conditions, while the secondary analysis populations are not reported in the registry record.

The record also does not report whether multiplicity was addressed, whether interim analyses occurred, or whether secondary outcomes were confirmatory. Death counts are not reported; null entries should not be interpreted as zero. Finally, the difference between the sponsor analysis and the unadjusted recalculation is material to interpretation. The sponsor estimate takes precedence, but the absence of model details prevents the recalculation from serving as an independent verification of that adjusted result.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: A Study Comparing Oral Calcitonin to Nasal Spray Calcitonin in Postmenopausal Osteoporotic Women — NCT00959764.
• Registry record last updated 2013-11-19; status last verified 2013-09.
• Results first posted 2013-11-19.
• 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

NCT00959764 is a completed randomized phase 3 parallel-group study of oral calcitonin, nasal calcitonin, and placebo in postmenopausal women with osteoporosis. Enrollment is listed as 565, and the planned study duration was 48 weeks. The primary endpoint was percent change from baseline in axial lumbar spine bone mineral density, measured by dual-energy X-ray absorptiometry.

The sponsor-reported primary ANCOVA estimate was 0.77, with a 95% confidence interval from 0.08 to 1.45 and p = 0.002. The registry classifies the analysis as non-inferiority or equivalence, but the margin is not reported. An unadjusted recalculation gives different comparison results; because the discrepancy flag is true, the sponsor’s analysis takes precedence when describing its result.

Two secondary CTx-1 outcomes are posted. Interpretation is constrained by unreported model details, multiplicity procedures, missing-data patterns, and the original sample-size justification. The primary outcome used last observation carried forward for some incomplete follow-up, but the number of such measurements is not reported in the registry record.