Evaluation of a Diagnostic to Identify Glucose-6-phosphate Dehydrogenase (G6PD) Deficiency in Brazil: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Evaluation of a Diagnostic to Identify Glucose-6-phosphate Dehydrogenase (G6PD) Deficiency in Brazil (NCT04033640), a non-randomized na study. The primary outcome is Sensitivity of SD Biosensor STANDARD G6PD Test for Identifying G6PD Deficient Individuals.

2. Clinical Question

3. Trial Design

The registry describes a completed, non-randomized, parallel diagnostic evaluation sponsored by PATH. The study title identifies Brazil as the setting. The recorded start date is 2019-06-27 and the primary completion date is 2019-12-31. The phase is listed as NA. Enrollment was 1754.

This was not a randomized treatment comparison. Participants in the diagnostic testing group provided whole blood and finger-stick capillary blood samples. At the clinic site, study staff performed the SD Biosensor STANDARD G6PD Test and the point-of-care HemoCue hemoglobin test on finger-stick samples. The record also lists the Pointe Scientific Test Kit and complete blood count as diagnostic tests. The available description of reference-laboratory procedures is incomplete, so further details of those procedures are not reported in the registry record.

A separate group of health workers received training on the SD Biosensor STANDARD G6PD Test from study team members experienced with G6PD diagnostics and the test. These participants were surveyed about packaging and label comprehension and interpretation of results. The record therefore describes more than one evaluative objective: test performance against reference methods, and usability or comprehension among intended users.

Masking is listed as none. The registry record does not describe a randomized allocation sequence, and the groups should not be interpreted as treatment arms in a controlled clinical trial. Their roles differ: diagnostic testing participants contributed blood samples, while health workers contributed information about use and interpretation. The study status is completed, but completion alone does not provide the outcome estimates. Numerical primary results and their confidence interval 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

Analysis population for the primary outcome, as defined in the registry: Participants with available data

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: Demographic information were not collected for the health worker participants in this study.

6. Primary Endpoint

The registry lists the primary outcome as Sensitivity of SD Biosensor STANDARD G6PD Test for Identifying G6PD Deficient Individuals, measured over All samples were collected on study day 1.

The primary endpoint is the sensitivity of the SD Biosensor STANDARD G6PD Test for identifying G6PD-deficient individuals. Sensitivity is the proportion of people who are positive according to the reference standard who are also identified as positive by the diagnostic test being evaluated. In this study, the Pointe Scientific assay defined the reference classification. The record defines a true positive as a participant with no more than 30% of normal G6PD activity in circulating venous blood according to that assay.

The endpoint is expressed as a percentage of participants. Its numerator is the number classified as positive by both the test and the reference definition; its denominator includes those true positives and participants who are reference-positive but test-negative. The latter are false negatives under this definition. The registry says sensitivity was calculated from venous and capillary blood samples and identifies the population as participants with available data. All samples were collected on study day 1.

Sensitivity is useful for describing how often a test detects the condition among people classified as having it by the chosen reference. It does not, by itself, describe performance among people without the condition, overall agreement, or the probability that a positive test represents deficiency. Those are distinct questions. The registry gives a 95% confidence interval as the planned dispersion measure, but neither the sensitivity estimate nor its interval is reported in the registry record.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does sensitivity measure?

In general, sensitivity is a conditional proportion: among individuals classified as having a target condition by a reference method, it measures the proportion that the test under evaluation identifies as positive. A false negative is someone classified as positive by the reference but negative on the test. Sensitivity therefore focuses on missed cases within the reference-positive group. It is different from specificity, which concerns reference-negative individuals, and from overall accuracy, which combines correct classifications across reference-positive and reference-negative groups.

For this study, the reference classification came from the Pointe Scientific assay. The registry defines deficiency as a positive result on that assay and specifies a true positive as no more than 30% of normal G6PD activity in circulating venous blood. The endpoint description says the calculation included results from venous and capillary samples. The numerical cross-classification counts needed to independently reproduce the sensitivity are not reported in the registry record.

Why use a reference method?

In general, diagnostic evaluations need a comparator that establishes the classification against which the test is judged. The reference method is not necessarily error-free; it supplies the operational definition used in the analysis. If the reference method misclassifies participants, estimated test performance can also be affected. The registry identifies the Pointe Scientific assay as the reference for G6PD deficiency, but does not report details that would permit a full assessment of its measurement properties or laboratory quality procedures.

Interpretation also depends on whether test results are assessed independently. If people reading one test know the result of another, their interpretation could be influenced. The record lists masking as none, but it does not describe whether result readers were blinded to other measurements or how discordant results were handled. These details are not reported in the registry record and should not be inferred.

How should a confidence interval be read?

In general, a confidence interval expresses statistical uncertainty around an estimate using a method with a stated coverage level. A narrow interval indicates greater precision than a wide interval, while both the point estimate and the interval are needed to understand the result. The interval does not tell whether the test is clinically useful, and it does not account automatically for every possible source of bias, such as limitations in the reference standard or participant selection.

The primary endpoint specifies a 95% confidence interval, but the estimate and interval limits are not reported in the registry record. It is therefore not possible to evaluate the precision of the sensitivity estimate from this record. The confidence interval method, including any handling of repeated specimens or dependence between venous and capillary measurements, is also not reported in the registry record.

Why can estimates vary across specimens?

In general, a diagnostic result may differ by specimen type because collection and measurement conditions can differ. Venous and capillary specimens are not interchangeable by assumption; comparing their results can address whether measurements align in the study setting. The registry lists median G6PD values measured by the SD Biosensor test for venous and capillary samples as a secondary outcome, and says the primary sensitivity used both types. It does not provide the estimates or the method used to combine or compare specimen-specific results.

When observations are paired within a participant, statistical methods should account for that pairing rather than treat the measurements as unrelated. This is a general methodological consideration, not a description of the analysis used in this study; the registry does not report how within-participant dependence was handled.

What do the planning assumptions mean?

The registry describes a sample-size planning illustration based on an observed effect, with a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. It says the effect to detect is a stated fraction of the observed effect. For outcomes other than time-to-event outcomes, such illustrations refer to participants per arm; time-to-event illustrations refer to total events. These are planning assumptions, not observed trial results and not observed power. The registry does not report the study’s own sample-size justification or enough detail to establish that this illustration was the study’s design calculation. Numerical planning outputs are not reported in the registry record.

In general, a planning calculation depends on assumptions about the effect to detect, variability, significance level, allocation, and analytic method. Changing an assumed effect changes the number of participants needed to achieve a chosen power under those assumptions. Such calculations do not show the probability that a completed study’s finding is true, and they cannot substitute for the observed estimate and its uncertainty. No further assumptions should be attributed to this study unless they are stated in the registry record.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists ten outcome measures in total and names several secondary outcomes. These include comparisons of G6PD test results with the Pointe Scientific assay, comparisons of hemoglobin measurements with HemoCue and complete blood count results, venous-capillary G6PD values, label comprehension, and accurate interpretation of results. The primary outcome is sensitivity of the SD Biosensor STANDARD G6PD Test for identifying G6PD-deficient individuals.

When a study examines multiple outcomes, the chance of at least one apparently unusual finding can increase if each result is interpreted without regard to the set of analyses. In general, a multiplicity plan may specify a hierarchy, adjustment, or other approach to control interpretation across outcomes. Whether such a plan was used is not reported in the registry record. The record does not report a detailed statistical analysis plan or an approach to multiplicity.

The primary analysis population is described as participants with available data. The registry does not report the number with available primary endpoint data, the number excluded from that analysis, or reasons for missing measurements. Missing-data handling, imputation, and sensitivity analyses are not reported in the registry record. It is therefore not possible to determine from the registry information whether missingness could have influenced the primary estimate.

The registry record also does not give enough detail to establish whether venous and capillary observations were analysed separately, jointly, or through a particular repeated-measures method. These details affect how the analysis population and uncertainty are interpreted. The description of available data identifies the intended population in broad terms, but does not replace reporting of participant flow, analysis denominators, and methods for incomplete or discordant measurements.

11. 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 0% reporting threshold.

The registry’s serious and other adverse event fields list the G6PD Diagnostic Testing group with 1736 participants at risk, 0 serious adverse events, 0 other adverse events, and 0 deaths. The Health Workers group has 0 listed at risk and 0 affected in each event field. No serious adverse event terms are listed.

These figures describe the registry entries, not a complete narrative of safety surveillance. In particular, the Health Workers group has 0 listed at risk, so its zero affected count should not be read as evidence that a defined group of participants was observed without events. The registry does not report additional detail about event collection, event definitions, follow-up duration, or how events were assessed.

Zero recorded events does not establish that no event could occur; it means the registry record lists none in the relevant fields. The serious adverse event counts should also be kept distinct from the study’s main purpose, which was diagnostic performance and user assessment. The record does not provide a treatment-effect safety comparison, and the non-randomized groups are not like-for-like control groups.

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

The registry record establishes what the study set out to evaluate: sensitivity of the SD Biosensor STANDARD G6PD Test against the Pointe Scientific assay, alongside other diagnostic comparisons and assessments of health-worker comprehension and result interpretation. It specifies that the primary endpoint concerns the percentage of reference-positive participants identified by the evaluated test and that the calculation used venous and capillary blood samples.

It does not report the numerical sensitivity estimate or its 95% confidence interval. As a result, the record does not permit a quantitative statement about how often the test identified reference-positive participants or how precisely that rate was estimated. The endpoint definition alone is not a result. Nor can it establish how performance differed by specimen type, because the relevant numerical comparisons are not reported in the registry record.

In general, sensitivity addresses test detection among people positive by the reference method. It does not by itself establish the test’s performance among reference-negative individuals, overall accuracy, or the likelihood that a positive result represents deficiency in a particular population. Those questions require additional measures and context. The listed secondary outcomes address some related comparisons, but their estimates are not reported in the registry record.

The study is also not a randomized comparison of clinical treatments. The diagnostic testing group and health-worker group had different roles, so their outcomes cannot be treated as outcomes from comparable intervention and control arms. The record describes a completed evaluation, but completion does not resolve the absence of posted numerical results or missing methodological detail.

Accordingly, the most supportable conclusion is descriptive: the registry defines a diagnostic accuracy question and identifies the reference standard and planned precision measure, but does not provide the primary result needed to characterize test performance. No treatment benefit or lack of benefit can be inferred from this diagnostic evaluation.

13. Why This Trial Matters Statistically

This study is statistically relevant because it separates a diagnostic accuracy question from a treatment comparison. Its primary endpoint is sensitivity against a defined reference assay, while other objectives concern hemoglobin measurement, specimen comparisons, and intended-user comprehension. Keeping these questions distinct is important: a test can be evaluated against a reference method, and its packaging or results can be evaluated with intended users, but these are different outcomes and require different interpretations.

The registry also makes explicit that G6PD deficiency was defined using the Pointe Scientific assay and a threshold of no more than 30% of normal activity in venous blood. A clear reference definition is central to interpreting sensitivity because the denominator is determined by who qualifies as reference-positive. The stated use of both venous and capillary samples also raises a meaningful methodological question about specimen-specific performance and dependence between measurements.

At the same time, the record illustrates why endpoint definitions must be accompanied by numerical estimates, analysis denominators, and uncertainty. Without the sensitivity and its confidence interval, readers cannot judge the observed performance or its precision. Without detailed missing-data and analysis procedures, they cannot assess the effect of incomplete measurements or how paired specimens were handled.

The non-randomized design is consistent with an evaluation of diagnostic performance rather than a controlled treatment trial, but the distinct roles of diagnostic testing participants and health workers should not be conflated. Finally, the adverse-event entries are limited counts and do not provide a broader safety analysis. The statistical value of the registry record therefore lies chiefly in defining the evaluation questions and clarifying what outcome information remains unavailable.

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 consequential limitation in interpreting the registry record is that it does not report the primary sensitivity estimate or its 95% confidence interval. The numbers of true positives and false negatives are also not reported, so the result cannot be independently reconstructed from the available description. Numerical results for the listed secondary outcomes are not reported in the registry record either.

The primary analysis population is described only as participants with available data. The record does not report the number contributing to the primary endpoint, participant flow, missingness by specimen type, or procedures for handling missing or discordant results. It also does not describe whether venous and capillary measurements were analysed separately or jointly, or how any within-participant dependence was handled.

The registry lists masking as none, but does not explain whether test readers knew reference results or whether reference assessment was independent of the evaluated test. Such procedural details can matter in diagnostic evaluations. Further detail about reference-laboratory measurement procedures is not reported in the registry record.

The record lists multiple outcomes but does not report a multiplicity approach or a detailed statistical analysis plan. The planning assumptions describe an illustration rather than establish the study’s own sample-size justification. Finally, the different participant roles mean the diagnostic testing and health-worker groups are not a like-for-like comparison. These limits restrict interpretation to the stated aims, definitions, and recorded counts rather than a quantitative assessment of diagnostic performance.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Evaluation of a Diagnostic to Identify Glucose-6-phosphate Dehydrogenase (G6PD) Deficiency in Brazil — NCT04033640.
• Registry record last updated 2021-11-11; status last verified 2021-09.
• Results first posted 2021-11-11.
• 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

NCT04033640 is a completed, non-randomized, parallel diagnostic evaluation in Brazil, sponsored by PATH. Enrollment was 1754. The study evaluated the SD Biosensor STANDARD G6PD Test against the Pointe Scientific assay and included hemoglobin comparisons, venous-capillary measurement comparisons, and health-worker assessments of packaging, labels, and result interpretation.

The primary outcome was sensitivity for identifying G6PD-deficient individuals, defined against the Pointe Scientific reference assay. The registry specifies a 95% confidence interval and says the analysis used venous and capillary samples, but the estimate and interval are not reported in the registry record. The record lists 1736 participants at risk in the G6PD Diagnostic Testing group, with 0 serious adverse events, 0 other adverse events, and 0 deaths; the Health Workers group has 0 listed at risk. Key details on analysis denominators, missing data, masking procedures, and multiplicity are not reported in the registry record.