Performance of an Investigational Blood Glucose Monitoring System in a Clinical Trial: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Performance of an Investigational Blood Glucose Monitoring System in a Clinical Trial (NCT01328717), a na na study. The primary outcome is Percent of Fingerstick Blood Glucose (BG) Results Within +/-20%(>=75 mg/dL) and Within +/- 15mg/dL (<75 mg/dL) of Laboratory Glucose Method.

2. Clinical Question

The primary endpoint concerns the agreement of system readings with laboratory glucose results. For reference results >=75 mg/dL, the stated agreement range is within +/-20%; for reference results <75 mg/dL, it is within +/-15 mg/dL. The registry describes the endpoint as a percentage of BG test results and states that test results were used to calculate the number meeting these criteria.

This is a device-performance question, not a comparison of clinical outcomes between treatment groups. The record describes 77 subjects who tested 2 test strip lots, yielding 154 test results. It does not report how many met the agreement criteria or the resulting percentage. Thus, the clinical question can be defined from the record, but its numerical answer cannot be determined from the information reported there.

3. Trial Design

The registry identifies NCT01328717 as a completed study titled “Performance of an Investigational Blood Glucose Monitoring System in a Clinical Trial.” The official title is “Clinical Trial Protocol for Contour Link System With Contour Sensor.” Ascensia Diabetes Care is listed as the sponsor. The condition is diabetes, and the intervention is a device: the Contour Link Investigational Blood Glucose Monitoring System.

The design is listed as single-group, with no masking. Allocation is recorded as NA, and phase is recorded as NA. There is one described group, “Intended Users of the System,” whose members used the investigational system. The record therefore does not describe a randomized comparison or a like-for-like control arm. Meter readings were instead compared with a laboratory reference method, the Yellow Springs Instrument (YSI) Analyzer.

Enrollment was 78. The study start and primary completion are both recorded as 2010-12, and the status is completed. The primary endpoint timeframe is 1 hour. The record says that one subject did not complete the study because of low blood sugar and rapidly changing glucose values. The remaining 77 subjects tested 2 test strip lots on the system, producing 154 available test results.

The registry record does not report further operational details such as the setting, participant selection criteria, or the order in which measurements were taken. It also does not describe a detailed statistical analysis plan. The single-group structure is important when interpreting the findings: the reference laboratory method supports an agreement assessment, but the study does not provide a randomized between-group estimate of comparative benefit or risk.

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: As a result of one subject's low blood sugar (and thus rapidly changing glucose values) during the study, one subject did not complete the study. The remaining 77 subjects tested 2 test strip lots on the system. 2x77(154) test results were available.

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 of Fingerstick Blood Glucose (BG) Results Within +/-20%(>=75 mg/dL) and Within +/- 15mg/dL (<75 mg/dL) of Laboratory Glucose Method, measured over 1 hour.

The primary endpoint is titled “Percent of Fingerstick Blood Glucose (BG) Results Within +/-20%(>=75 mg/dL) and Within +/- 15mg/dL (<75 mg/dL) of Laboratory Glucose Method.” It addresses whether fingerstick results from the investigational blood glucose meter were sufficiently close to the laboratory reference results according to two reference-value-dependent criteria. When the reference glucose result was >=75 mg/dL, a meter result counted as within the stated range if it was within +/-20%. When the reference result was <75 mg/dL, the stated range was within +/-15 mg/dL.

This type of endpoint is used to describe agreement between a measurement device and a reference method. It focuses on the frequency of readings meeting prespecified tolerances, rather than on a change in a participant’s health or a difference between treatment groups. The registry lists the unit as a percentage of BG test results, while its description says the results were used to calculate the number meeting the criteria. The result itself is not reported in the registry record.

The endpoint timeframe is 1 hour. The record identifies the YSI Analyzer as the reference method and reports 154 available test results from 77 subjects who tested 2 test strip lots. It does not explain whether the endpoint was calculated separately by lot or pooled across lots, nor does it provide a confidence interval or other measure of uncertainty.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The primary endpoint uses a threshold-based agreement rule. Each eligible fingerstick result is considered relative to its corresponding laboratory reference result. The acceptable difference depends on the reference value: the endpoint specifies within +/-20% for reference BG results >=75 mg/dL and within +/-15 mg/dL for reference BG results <75 mg/dL. The endpoint title calls the measure a percent, while the unit is “Percentage of BG test results”; the description says the results were used to calculate the number within the stated ranges. No numerator or resulting percentage is reported in the registry record.

What does a threshold-based agreement measure describe?

In general, a threshold-based agreement measure summarizes the share or count of measurements that fall within a predefined tolerance of a reference. It offers a compact way to express how often a device reading is near the reference according to the chosen rules. It does not, by itself, describe the direction or size of errors for readings outside the tolerance. A full understanding can also depend on the distribution of differences and whether those differences vary across the measurement range. Those additional summaries are not reported in the registry record.

The choice of thresholds is part of the endpoint definition. Results with reference values on different sides of the stated cutoff are assessed using different forms of tolerance: a percentage difference at or above the cutoff and an absolute difference below it. The record does not provide further rationale for selecting these rules or explain whether additional performance criteria were specified.

Why does the count of test results differ from the count of subjects?

The record reports 77 subjects with available results, each testing 2 test strip lots, for 154 available test results. These are repeated measurements within subjects, not a report of 154 separate participants. In general, measurements from the same participant can be correlated: a person’s readings may share biological and measurement characteristics. If correlation is present, an analysis that assumes every reading is independent can understate uncertainty. The record does not say whether the analysis used a method that accounted for repeated measurements, and it does not report an analysis at the participant level or at the test-result level.

Testing 2 lots also raises a question about how lot-specific information was handled. An analysis might report results by lot, combine them, or use a model that accounts for lot structure; the registry record does not report which approach, if any, was used. No lot-specific outcome values are given.

What can a confidence interval add?

In general, a confidence interval communicates the precision of an estimated quantity by giving a range of values compatible with the data under the method’s assumptions. For an agreement percentage, it can help readers distinguish a precise estimate from one that remains uncertain. The interval’s interpretation depends on how the observations were sampled and analyzed, including whether repeated measurements were handled appropriately. The registry record does not report an endpoint estimate or confidence interval, so the precision of the endpoint cannot be assessed from this record.

The record also reports zero analyses. It therefore does not document an inferential method, a model, an adjustment strategy, or a hypothesis test for the primary outcome. It is not possible to identify whether an analysis was planned or performed outside the registry information.

How should planning assumptions be read?

The stated sample-size assumptions describe a planning illustration based on an observed effect, rather than the trial’s own design. The illustration assumes a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; it considers an effect defined as a stated fraction of the observed effect. These are planning assumptions, not reported features of the single-group study, which has allocation recorded as NA and no separate randomized arms.

In general, a sample-size calculation estimates how many participants would be needed under specified assumptions to detect a target effect with a chosen probability of rejecting a null hypothesis when that effect is present. The answer depends on the assumed effect, variability, significance level, power target, allocation, and analysis method. The registry record does not report the observed primary endpoint effect or numerical planning outputs, so those calculations cannot be evaluated from the record. Planning counts do not measure the probability that this completed study detected an effect and should not be called observed power.

Over-dispersion is a general concern when the variability in observed counts or measurements is greater than a model assumes. In repeated-measurement data, unmodeled clustering can contribute to underestimated uncertainty. The record does not report a model or variability estimate, so it is not possible to determine whether over-dispersion was relevant or addressed in an analysis of these results.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists 2 outcome measures and 0 analyses. The primary endpoint is the agreement measure for fingerstick BG results. A listed secondary outcome is the number of subjects rated as <=3 performing basic meter tasks in a labeling evaluation. The record does not report results for that secondary outcome or provide analyses for either outcome.

Multiplicity arises in general when several outcomes, time points, subgroups, or statistical tests are examined. If multiple tests are used to support confirmatory conclusions, the chance of at least one false-positive finding can increase unless the analysis plan addresses that issue. The registry record does not report whether the outcomes were part of a prespecified testing hierarchy, whether any multiplicity adjustment was planned, or whether additional exploratory outcomes or analyses were considered. No multiplicity procedure can therefore be identified.

The record says one subject did not complete the study because of low blood sugar and rapidly changing glucose values. The remaining 77 subjects tested 2 test strip lots, and 154 test results were available. This reports the number of available results, but does not specify the complete analysis population or state whether all available results were included in an endpoint analysis. It does not report how missing readings, incomplete subject data, or any excluded observations were handled.

In general, missing-data handling can affect the interpretation of an estimate, particularly when the reason data are missing may be related to the measurement or the outcome. The stated reason for one noncompletion is part of the study description, but the record does not give enough information to assess the impact of that event on an endpoint estimate. It also does not report sensitivity analyses or an intention-to-measure, per-protocol, or other analysis population. Details of the statistical analysis plan and population definitions are not reported in the registry record.

11. 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 2% reporting threshold.

Most frequent non-serious adverse event terms

The registry lists 78 subjects at risk for the serious adverse event summary and reports 0 subjects affected by a serious adverse event. For other adverse events, 78 subjects were at risk and 2 were affected. These are subject counts, not a measure of event severity beyond the categories used in the record. The registry does not list any top serious adverse event terms.

The deaths field is null. A null entry is not the same as a reported count of zero; the number of deaths is not reported in the registry record. The record also does not provide definitions for the serious and other adverse event categories, details about collection or follow-up, or descriptions of the 2 other events.

These counts should be read as the adverse event information recorded for the study, not as a complete comparison of risks between interventions. The study has one described group and no randomized control arm. The counts alone do not establish whether an event was caused by the device, and no causal attribution is reported. They also do not quantify how precisely the event frequency is estimated.

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

The record establishes the intended measurement question: whether fingerstick blood glucose results from the investigational system fell within specified tolerances of the YSI Analyzer reference results. It also reports that the 77 subjects who completed the study tested 2 test strip lots, yielding 154 available test results. The primary endpoint’s criteria and one-hour timeframe are defined.

What the record does not provide is the observed number or percentage of results meeting those criteria. It reports 0 analyses, and does not give an endpoint estimate, confidence interval, or test result. Readers therefore cannot use the registry record to determine the observed level of agreement or its statistical precision.

Even if an agreement estimate were available, its meaning would be specific to this measurement comparison and the endpoint rules. It would not, on its own, answer questions about clinical outcomes, establish a difference from another device, or show a randomized treatment effect. The study has a single group and uses a laboratory reference rather than a control arm.

The counts of subjects and tests also represent different units of observation. The 154 test results were produced by 77 subjects testing 2 lots, so the results may not be statistically independent. The record does not describe how this structure was handled. This is relevant to precision and to any inference based on the test-result count.

The report of one noncompletion provides a reason related to low blood sugar and rapidly changing glucose values, but the registry does not explain how that subject’s data were handled or whether other missing data occurred. Likewise, the adverse event counts do not supply a comparative safety estimate. Overall, the registry record describes the study and its endpoint, but does not report outcome analyses from which a numerical performance conclusion can be drawn.

13. Why This Trial Matters Statistically

This study illustrates a common statistical distinction in device evaluation: agreement with a reference measurement is not the same question as comparative effectiveness between randomized groups. The endpoint is based on whether meter readings fall within specified tolerances of laboratory values. The reference method anchors the measurement comparison, while the single-group design means there is no separate randomized control arm.

The record also makes clear why the unit of analysis matters. It reports 154 available test results from 77 subjects who tested 2 lots. Multiple readings from the same person can carry shared information, so the number of readings alone does not establish the amount of independent information available. Statistical methods should reflect the structure of repeated measurements if an analysis is performed; the registry record does not report whether or how this was done.

A further statistical lesson is the distinction between defining an endpoint and reporting its result. The registry gives threshold rules, a timeframe, and the available test count, but does not provide the numerator, percentage, or uncertainty estimate. Without those quantities, the endpoint cannot be interpreted quantitatively.

Finally, planning calculations and study results serve different purposes. The record describes a planning illustration based on an observed effect and assumptions for a two-arm comparison, but the trial itself is single-group. Planning numbers do not substitute for an observed endpoint analysis and are not evidence of the study’s achieved power. Separating these concepts helps prevent a design calculation from being mistaken for a finding.

14. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The most important limitation of the registry record for statistical interpretation is that it does not report the primary endpoint result. Although the endpoint rules and 154 available test results are described, no number or percentage of readings meeting the criteria is given. There is also no reported confidence interval, model, or inferential analysis; the record lists 0 analyses.

The study is a single-group, unmasked design, with allocation recorded as NA. It does not include a randomized control group, so it cannot provide a randomized comparison between interventions. Its laboratory reference comparison addresses a different question: agreement under the endpoint’s thresholds. The record does not report details about how measurements were paired with the reference method or how observations were ordered.

The 154 test results came from 77 subjects testing 2 test strip lots. Because readings are clustered within subjects and may also relate to lot, treating each test as independent could affect uncertainty. The registry does not report whether results were pooled across lots, analyzed separately, or modeled to account for repeated observations. It also does not explain whether all available results entered the primary endpoint analysis.

One subject did not complete the study because of low blood sugar and rapidly changing glucose values. The record gives no further account of this participant’s measurements, missing-data handling, or sensitivity analyses. It does not report detailed eligibility criteria, the analysis populations, an analysis plan, or multiplicity procedures. These omissions limit assessment of bias, precision, and the robustness of any result. The adverse event record is similarly limited: it gives counts but no event descriptions, category definitions, or reported death count.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Performance of an Investigational Blood Glucose Monitoring System in a Clinical Trial — NCT01328717.
• Registry record last updated 2016-02-29; status last verified 2016-01.
• Results first posted 2012-05-31.
• 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.

18. Record Summary

NCT01328717 is a completed, single-group study of the Contour Link Investigational Blood Glucose Monitoring System in subjects with diabetes. The record lists 78 enrolled subjects, no masking, and allocation as NA. Its primary endpoint assesses whether fingerstick BG results fall within specified tolerances of the YSI Analyzer reference method during a one-hour timeframe.

One subject did not complete the study because of low blood sugar and rapidly changing glucose values. The remaining 77 subjects tested 2 test strip lots, with 154 test results available. The registry record does not report the primary endpoint result or an analysis; it lists 2 outcome measures and 0 analyses. For safety, 0 subjects are reported as affected by serious adverse events and 2 by other adverse events among 78 at risk; the death count is not reported in the registry record.