Pharmacogenomic Testing in Primary Care: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Pharmacogenomic Testing in Primary Care (NCT03270891), a randomized na study. The primary outcome is Count of Participants Prescribed Antidepressant Medication.

2. Clinical Question

This is a question about both clinical workflow and patient-level outcomes. Making test information available sooner could change prescribing decisions, while the pilot also addresses whether testing and result use can be carried out in primary care. The registry record describes the study as a 6-month randomized, wait-listed, controlled pilot trial. Its stated objective includes feasibility, acceptability, and preliminary effectiveness. The primary outcome definition focuses on prescribing that is not contraindicated, rather than directly measuring whether patients’ symptoms improved. Changes in symptom severity and medication adherence were listed as secondary outcomes. The registry record does not report results for these outcomes, so the clinical question can be described, but its answer cannot be inferred from the outcome definitions alone.

3. Trial Design

The study was a randomized, parallel-group trial with no masking. It enrolled 75 participants and is listed as completed. The registry record gives the start date as 2017-12-01 and the primary completion date as 2021-09-05. The study was described as a 6-month, wait-listed, controlled pilot conducted in 6 University of Michigan department of family medicine clinics.

In the intervention arm, physicians were to receive pharmacogenomic test results as soon as the results were available after enrollment. In the delayed-intervention control arm, results were to be made available 3 months after enrollment. This wait-listed arrangement means the contrast concerns the timing of access to test results; the control condition was not described as permanent withholding. The registry record does not describe the randomization sequence or allocation concealment, so those procedures cannot be characterized further.

The described clinician population consisted of physicians practicing at a University of Michigan department of family medicine clinical site who were willing to use pharmacogenomic results to manage antidepressant use. Patients of those physicians could enroll if they were adults with a new prescription for a target antidepressant. The registry record does not give additional eligibility criteria, recruitment procedures, or participant-flow details.

The intervention is classified as a diagnostic test. The trial was not masked, so participants and clinicians were not described as blinded to assignment. The record identifies the condition as antidepressants causing adverse effects in therapeutic use, but this does not by itself establish the clinical history or outcome for any individual participant. The phase is listed as not applicable. A sample-size justification, the planned statistical model, and details of any interim analyses 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: Per protocol, only control arm patients were to have a 9 month assessment.

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: 1 participant was withdrawn by the study staff and PI after baseline data was collected but before receiving the intervention.

6. Primary Endpoint

The registry lists the primary outcome as Count of Participants Prescribed Antidepressant Medication, measured over Baseline, 3 months, 6 months, and 9 months.

The primary endpoint is the count of participants prescribed antidepressant medication that was not contraindicated. It was scheduled for assessment at baseline, 3 months, 6 months, and 9 months. The registry record specifies that, per protocol, only control-arm patients were to have a 9-month assessment. The endpoint is therefore a repeated measure of a prescribing status, rather than a direct measure of symptom change or medication adherence.

In general, a prescribing endpoint can help describe whether treatment decisions align with a defined medication-safety criterion. In this trial, the criterion is stated as medication that is not contraindicated. A count indicates how many participants meet that description at an assessment; interpreting a count as a comparison between arms also requires information about the number assessed in each arm and the analysis used. Endpoint-specific denominators, estimates, uncertainty intervals, and statistical comparisons are not reported in the registry record.

The timing matters for interpretation. Repeated assessments can describe how the endpoint changes over follow-up, but the 9-month observation was planned only for the control arm. That asymmetric follow-up means a 9-month comparison between arms cannot be assumed from the endpoint schedule. The registry record does not report the observed endpoint values, so it is not possible to describe the direction or size of any change. The endpoint’s definition clarifies what was intended to be measured; it does not establish that prescribing changed or that any change was caused by access to test results.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry record describes a randomized parallel-group design without masking. In general, randomization is intended to make the assigned groups comparable, on average, with respect to measured and unmeasured characteristics. It does not guarantee identical groups, especially in a pilot trial, and it does not remove the need to report how assignment was generated and implemented. The randomization method and allocation concealment are not reported in the registry record.

What does the prescribing count represent?

The primary measure is a count of participants prescribed antidepressant medication that was not contraindicated. A count is a frequency: it states how many participants meet the endpoint definition at an assessment. In general, a count alone does not reveal the proportion of an arm meeting the definition unless the corresponding denominator is also known. Nor does a count by itself describe the size or precision of a difference between randomized groups. The registry record does not report the primary outcome counts or endpoint-specific denominators.

The endpoint is scheduled at baseline, 3 months, 6 months, and 9 months. Repeated measurements can be analyzed in several ways, including comparisons at individual follow-up times or models that account for measurements from the same participant being related. These approaches answer somewhat different questions. The planned statistical approach, the contrast designated as primary, and the handling of within-participant correlation are not reported in the registry record. No particular model should therefore be attributed to this trial.

Why do assessment times matter?

Assessment timing defines which observations can contribute to each comparison. The record states that only control-arm patients were to have a 9-month assessment. As a general statistical matter, when follow-up schedules differ, a direct comparison at the unmatched time point may not be available. It is also important to distinguish a within-arm change over time from a between-arm contrast: a change in one group alone does not establish that the groups differ. The registry record does not state which contrast or time point was used for the primary inference.

What would an uncertainty interval add?

In general, a confidence interval describes the range of effect estimates compatible with the data and the assumptions of the analysis. It conveys precision, whereas a point estimate alone does not. A narrow interval indicates greater precision than a wide interval, but neither alone determines clinical importance. No primary endpoint estimate or confidence interval is reported in the registry record, so the precision of any prescribing effect cannot be assessed.

What does the pilot design establish?

The record calls this a pilot trial and describes feasibility, acceptability, and preliminary effectiveness as aims. In general, pilot studies can examine whether recruitment, testing, result delivery, and data collection are workable, while also gathering preliminary outcome information. A pilot label does not reveal what outcomes were achieved or whether a later definitive study was planned. The registry record does not report a sample-size justification or the criteria used to assess feasibility and acceptability.

The sample-size assumptions described in the record refer to a planning illustration based on an observed effect, with equal allocation and a two-sided significance level of 0.05. Such a planning illustration is not an observed-power result and does not establish that the trial was designed to detect a particular effect. The numerical planning values and the effect estimate underlying that illustration are not reported in the registry record. No sample-size conclusion can be drawn from the assumptions alone.

How should event counts be interpreted?

For a binary event, a count records whether participants experienced the event, while the number at risk supplies the context for that count. In general, the amount of information in a count depends on both the event frequency and the number observed; zero observed events do not prove that the underlying event risk is zero. The safety summary reports event counts and at-risk totals, but the record does not provide inferential estimates for those safety outcomes. These safety counts should not be confused with the primary prescribing endpoint.

Finally, statistical methods are part of how results are interpreted, not a substitute for reporting the results themselves. The analysis population, missing-data approach, multiplicity procedure, and inferential model are not reported in the registry record. Without those details and the endpoint estimates, it is not possible to assess the primary analysis or its uncertainty.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists 5 outcome measures and 5 analyses. It names three secondary outcomes: change in PHQ-8 symptom severity, change in GAD-7 symptom severity, and change in ARMS scores. The primary endpoint is measured at baseline, 3 months, 6 months, and 9 months, with the record noting that only control-arm patients were to have the 9-month assessment. The relationship between the listed analyses and each outcome or assessment time is not described in detail.

In general, testing multiple outcomes or time points creates multiple opportunities to observe a nominally notable result by chance. A prespecified hierarchy or adjustment procedure can address that issue, but the registry record does not report whether multiplicity was controlled, which outcome was treated as the main inferential focus, or how time-point comparisons were prioritized. No adjustment should be assumed.

Missing-data handling is also not reported. The record does not give participant flow by assessment, the number of observations missing for each outcome, reasons for missingness, or the method used to analyze incomplete follow-up. In general, missing outcomes can affect both the precision and interpretation of an estimate, particularly when availability differs by arm or over time. The planned 9-month assessment for control-arm patients only is an aspect of the design, not a missing-data method.

The endpoint description refers to a per-protocol assessment schedule, but the registry record does not define the full analysis population or explain whether results were analyzed according to randomized assignment, protocol adherence, or another rule. It also does not report sensitivity analyses. These omissions limit assessment of how the analysis addressed multiplicity, incomplete follow-up, and departures from the assigned intervention timing.

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.

In the delayed-intervention control arm, the registry record lists 26 participants at risk for serious adverse events and 0 affected; it also lists 26 at risk for other adverse events and 0 affected, with 0 deaths. In the PGx Test arm, it lists 25 participants at risk for serious adverse events and 0 affected; it also lists 25 at risk for other adverse events and 0 affected, with 0 deaths. No top serious adverse-event terms are listed.

These figures are event counts within the safety groups recorded as at risk. In general, zero observed events means that none were recorded in the listed group during the period captured by the safety summary. It does not establish that the true risk is zero, or that an event could not occur in a larger or differently followed population. The record does not provide additional detail about the safety observation period, event ascertainment, or the reason the at-risk denominators differ from the overall enrollment of 75.

The counts describe safety reporting only. They do not answer the primary prescribing question or establish an effect on symptoms or adherence. No causal or broader safety conclusion should be drawn from these counts alone.

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

The registry record describes a randomized pilot comparing earlier with delayed availability of pharmacogenomic test results for physicians managing antidepressant use in primary care. It defines a prescribing endpoint based on the count of participants receiving antidepressant medication that was not contraindicated, and identifies symptom severity and adherence as secondary outcomes. It also reports arm-specific safety counts with no serious adverse events, other adverse events, or deaths recorded among the listed participants at risk.

Those details describe the study’s question and the limited results available, but they do not show whether prescribing differed between arms. The registry record does not report observed values for the primary endpoint, estimates of between-arm effects, or uncertainty intervals. It also does not report results for the secondary outcomes. Therefore, neither an effect nor the absence of an effect can be established from these details. A missing result is not equivalent to a result showing no difference.

The study’s pilot designation and stated aims encompass feasibility, acceptability, and preliminary effectiveness. The design and setting indicate what was intended to be examined, but the registry record does not report findings on feasibility or acceptability. It also does not give enough information to judge whether findings from the participating clinics and eligible patients would apply in other settings.

In general, a trial result is interpreted by considering the estimate, its precision, the population studied, and the analysis used. Here, the endpoint definitions and safety counts are available, while key outcome estimates and analysis details are not reported in the registry record. The responsible interpretation is consequently limited: the trial was designed to evaluate timely access to test results and related outcomes, but the posted registry information does not establish the size, direction, or certainty of any effect.

13. Why This Trial Matters Statistically

This trial raises a statistical question about how to evaluate a diagnostic test when its intended role is to inform prescribing decisions. The primary endpoint is not simply whether a test was performed; it is a repeated count of participants receiving antidepressant medication described as not contraindicated. That distinction matters because test availability, clinician use of results, and prescribing outcomes are different stages in the pathway from testing to a measured clinical outcome.

The wait-listed design also makes timing central to the comparison. Physicians in the intervention arm were to receive results as soon as available, while results in the control arm were to be released 3 months after enrollment. The endpoint schedule includes baseline and follow-up assessments, but the record notes that only control-arm patients were to have a 9-month assessment. Interpreting repeated outcomes therefore depends on clearly specifying which time point and contrast answer the primary question.

The trial is described as a pilot addressing feasibility, acceptability, and preliminary effectiveness. Statistically, those aims are related but not interchangeable. Feasibility concerns whether study procedures can be carried out; acceptability concerns whether participants and clinicians find them acceptable; effectiveness concerns outcome differences. The registry record does not report results for those aims, so their contribution cannot be assessed from the information recorded.

The record also illustrates why outcome definitions and safety counts are not enough to evaluate an intervention. Interpretable effect estimates require observed outcomes, denominators, an analysis plan, and information about missing follow-up. Those details are not reported for the primary endpoint. The study’s design provides a framework for asking a clinically relevant question, while the available registry information leaves its statistical answer unresolved.

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 registry record does not report observed results for the primary prescribing endpoint or the listed secondary outcomes. It therefore does not permit assessment of effect direction, magnitude, or precision. The primary endpoint is assessed repeatedly, and only control-arm patients were to have a 9-month assessment; the record does not describe the principal time-point comparison or how the unequal schedule was addressed analytically.

The overall enrollment is 75, while the safety summary lists 26 participants at risk in the delayed-intervention control arm and 25 in the PGx Test arm. The registry record does not reconcile these figures or provide a participant-flow account. It also does not report outcome-specific denominators, missingness by arm or assessment, reasons for missing data, or the methods used to handle incomplete observations.

Further trial details that would help evaluate the evidence are not reported in the registry record. These include the randomization procedure, allocation concealment, the statistical model, the analysis population, multiplicity handling, adherence to assigned result timing, and a sample-size justification. The record also does not specify how feasibility and acceptability were measured or report their findings.

The study took place in 6 University of Michigan department of family medicine clinics and enrolled patients with a new prescription for a target antidepressant from physicians willing to use pharmacogenomic results. The registry record does not provide enough information to characterize how participants were recruited or to judge representativeness beyond that setting and eligibility description. These limitations concern what can be evaluated from the registry record; they do not by themselves establish that the trial procedures or outcomes were inadequate.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Pharmacogenomic Testing in Primary Care — NCT03270891.
• Registry record last updated 2023-04-11; status last verified 2023-03.
• Results first posted 2023-04-11.
• 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

NCT03270891 is a completed randomized, unmasked, parallel-group pilot trial sponsored by the University of Michigan. It enrolled 75 participants and studied pharmacogenomic testing in primary care. Physicians in the intervention arm were to receive test results as soon as available after enrollment; control-arm physicians were to receive results 3 months after enrollment. The study was conducted in 6 department of family medicine clinics and was described as a 6-month wait-listed controlled pilot.

The primary endpoint was the count of participants prescribed antidepressant medication that was not contraindicated, assessed at baseline, 3 months, 6 months, and 9 months. The registry record notes that only control-arm patients were to have a 9-month assessment. Secondary outcomes included PHQ-8 symptom severity, GAD-7 symptom severity, and ARMS scores. Outcome results and statistical analysis details are not reported in the registry record.

The safety summary lists zero serious adverse events, zero other adverse events, and zero deaths in each arm. The arm-specific at-risk counts are 26 for the delayed-intervention control arm and 25 for the PGx Test arm.