Medication Optimization Using Pharmacogenetic Testing and the G-DIG to Reduce Polypharmacy in a Mental Health Population: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Medication Optimization Using Pharmacogenetic Testing and the G-DIG to Reduce Polypharmacy in a Mental Health Population (NCT03468309), a registered phase not stated study. The primary outcome is Clinical Global Impression.

2. Clinical Question

The study asks whether providing pharmacogenetic testing together with the Genomind Drug Interaction Guide (G-DIG) is associated with changes in medication use and patient-reported or clinician-assessed outcomes in a mental health population. It is described as a 12-week, open-label, naturalistic study. The registry record does not report numerical results for the primary outcome, so the size and precision of any change cannot be assessed from the record.

3. Trial Design

The registry describes a 12-week, open-label, naturalistic study of pharmacogenetic testing and a computerized decision tool. Enrollment was 53, and the record lists the study status as completed. The recorded study start date is 2018-01-01, and the primary completion date is 2019-04-01. The sponsor is the Seattle Institute for Biomedical and Clinical Research.

The intervention combines the Genecept Assay with the G-DIG decision tool. Participants were veterans prescribed 5 or more medications under the eligibility description. They were required to have at least two medications for a mental health diagnosis, with an alternative allowance for one mental health medication and another medication for its side effects. These criteria identify a population with medication use relevant to the study question; they do not by themselves establish that participants had a particular level of symptom severity or a specific medication pattern beyond the stated eligibility.

The record does not state an allocation method or describe a control arm. It explicitly characterizes the study as not being a randomized comparison, and the arms are not a like-for-like control comparison. Consequently, interpretation rests on outcomes observed in the study population rather than a randomized contrast between intervention and control groups. The record does not report details of how providers used the test or tool, whether medication changes were required, or how treatment decisions were standardized.

The primary outcome is Clinical Global Impression at baseline and 12-weeks. Listed secondary outcomes are number of psychiatric medications, Patient Health Questionnaire-9 (PHQ-9), and Generalized Anxiety Scale-7 (GAD-7). The registry record does not report the results or detailed analysis procedures for these outcomes. It also does not report the full protocol, sample-size justification, or other design details that would clarify how the study was planned and conducted.

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:

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: Participants as described in data tables.

6. Primary Endpoint

The registry lists the primary outcome as Clinical Global Impression, measured over Baseline and 12-weeks.

The primary endpoint is Clinical Global Impression, described as a measure in units on a scale and assessed at baseline and 12-weeks. Its stated range is 1-6, with 1 indicating less symptomology and 6 indicating high symptomology. Thus, the direction of the scale is specified: higher values correspond to greater symptomology as described in the registry.

In general, a global clinical impression measure can offer a concise overall assessment of clinical status, rather than focusing on a single symptom or behavior. That broad scope can make it useful alongside more specific measures, although interpretation depends on how the scale is administered and who provides the rating. The registry record does not describe the rating procedure, identify the rater, or give further details about how the score was derived.

The endpoint is recorded at baseline and at 12-weeks, which gives the study two stated assessment times. However, the registry record does not report the mean score or standard deviation at either time, a change estimate, or an uncertainty interval. It therefore is not possible to determine from the record whether scores changed, how large any change was, or how precisely it was estimated. The scale's range and direction explain what the measurement represents, but they are not evidence of an outcome.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The registry identifies Clinical Global Impression as a continuous outcome summarized by a mean and standard deviation. The record does not provide the numerical mean or standard deviation at either assessment time, nor does it report an analysis estimate. The concepts below explain how such statistics are generally interpreted; they are not claims about analyses that were performed in this study.

What does a mean describe?

In general, a mean is the arithmetic average of the observed values in a group. It gives a central summary, but does not describe how widely individual observations vary. For this endpoint, the scale direction matters when interpreting a mean: the registry states that 1 corresponds to less symptomology and 6 to high symptomology. A mean alone also does not show whether change was similar across participants or driven by a subset of observations.

What does a standard deviation describe?

In general, a standard deviation describes the spread of observations around their mean. A smaller standard deviation indicates observations are more tightly clustered around the mean, while a larger one indicates more variability. It is not the same as the standard error, which describes uncertainty in an estimated mean. The registry names standard deviation as the dispersion measure for the primary outcome but does not report its value.

How should a change estimate be read?

In general, a change estimate compares measurements over time or contrasts groups. Its interpretation depends on the defined calculation, the direction of the scale, and the design. A baseline-to-follow-up change in a single group describes change over time among participants with available measurements; by itself, it does not isolate the effect of an intervention. A randomized comparison can support a different kind of inference because allocation is intended to make groups comparable, but the registry does not describe a randomized comparison for this study.

What does a confidence interval add?

In general, a confidence interval gives a range of values compatible with the data and the statistical method used, expressing uncertainty around an estimate. It helps readers distinguish a precise estimate from one with substantial uncertainty. It does not show the probability that a particular result is true, and its interpretation depends on the model and assumptions. The registry record does not report a confidence interval for the primary outcome.

What does the planning illustration represent?

The sample-size material is described as a planning illustration computed from the observed effect, not as the trial's own design calculation. It uses a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation; the effect to detect is a stated fraction of the observed effect. The record also specifies that time-to-event calculations give total events, while other calculations give participants per arm.

These assumptions describe an illustration, not the actual study design. In particular, an equal-allocation, two-arm calculation should not be mistaken for a randomized control comparison in a study that the registry describes as nonrandomized and without a like-for-like control. The observed effect on which the illustration is based is not reported in the registry record, so its numerical planning implications cannot be assessed from the record. Such calculations are not observed power: they do not show the probability that the completed study detected an effect.

Why can variability and model choice matter?

In general, estimates from continuous outcomes are affected by the spread of observations and by the number of observations available for analysis. Greater variability can make an estimate less precise. The assumptions used for a statistical model also matter; a model should reflect the design, outcome distribution, and dependence among measurements. The registry record does not state the model or test used for Clinical Global Impression, or whether repeated measurements were analyzed jointly.

The outcome is measured at baseline and 12-weeks, so an analysis could in principle account for both measurements, or could analyze change from baseline. Those approaches can answer related but not identical questions and rely on different modeling choices. The record does not specify which approach was used. It also does not report confidence intervals, test statistics, or a detailed statistical analysis plan. Therefore, general statistical definitions can guide interpretation of any reported estimates, but cannot fill in the missing analysis details or results.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists Clinical Global Impression as the primary outcome and names three secondary outcomes: number of psychiatric medications, Patient Health Questionnaire-9 (PHQ-9), and Generalized Anxiety Scale-7 (GAD-7). It does not report outcome results or describe a multiplicity strategy. When several outcomes are tested, the chance of at least one apparently notable result can increase if each is interpreted without accounting for the set of tests. In general, an analysis plan may designate a primary outcome or adjust inference across multiple tests, but the registry record does not state whether such procedures were used.

The record does not describe missing-data amounts or methods. It does not report how participants with absent baseline or follow-up measurements were handled, whether any imputation was performed, or what assumptions were used. In general, missing outcomes can affect both the estimate and its uncertainty, especially if participants with missing measurements differ from those with observed measurements. Without counts and handling details, the extent and possible influence of missing data cannot be judged.

The analysis population is also not described. The enrollment was 53, but the record does not say how many participants contributed data to each outcome analysis or whether analyses used all enrolled participants, participants with complete measurements, or another defined group. Enrollment and analyzed sample size are not interchangeable. The registry record does not report a statistical analysis plan specifying populations, exclusions, outcome definitions beyond the listed descriptions, or adjustments. These omissions limit assessment of how fully the analysis followed a prespecified approach and how robust the findings would be to alternative choices.

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 lists 53 participants at risk for serious adverse events and 0 affected. For other adverse events, it also lists 53 at risk and 0 affected. Deaths are listed as 0, and no serious-event terms are listed. These are the safety counts recorded for the intervention group.

A count of zero means that no events were recorded in the relevant category in the registry entry; it does not establish that an event could never occur or that risk is absent. In general, safety counts are most informative when accompanied by clear definitions, the period and methods of collection, and a comparison group. The registry record does not report those additional details, including how events were solicited or classified.

Because no like-for-like control comparison is described, the counts do not provide a comparative estimate of risk. They should not be treated as evidence that the intervention reduced adverse events, or as a precise estimate of the frequency of rare events. The entries describe what was reported in the registry, while the record leaves the event definitions and collection procedures unspecified.

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

The registry record establishes that a completed, open-label, naturalistic study enrolled 53 participants to assess pharmacogenetic testing with the G-DIG decision tool in a mental health population prescribed multiple medications. It identifies Clinical Global Impression as the primary outcome, assessed at baseline and 12-weeks, and lists medication count, PHQ-9, and GAD-7 as secondary outcomes. Those details clarify the study's intended question and measurement framework.

The record does not report numerical results for these outcomes. It therefore does not show the direction or magnitude of changes in clinical impression, medication count, depression questionnaire scores, or anxiety questionnaire scores. Nor does it provide uncertainty estimates with which to assess precision. The outcome names and scale description should not be interpreted as findings.

The design also limits what could be concluded from change over time. The study is described as open-label and naturalistic, and the registry does not describe a randomized comparator or allocation method. In general, without a suitable comparison group, changes cannot be confidently attributed to the intervention rather than other influences occurring during follow-up. The record also does not report the analysis population, missing-data handling, or detailed statistical plan.

The safety entries report no participants affected by serious or other adverse events and no deaths. Those counts are descriptive registry entries, not comparative evidence, and the event definitions and ascertainment procedures are not reported. Overall, the record supports a description of the study's aims, population, intervention, and selected outcomes, but it does not support a quantitative conclusion about outcome changes or their precision.

13. Why This Trial Matters Statistically

This study raises a statistically important distinction between evaluating a combined decision-support intervention and estimating its effect. The intervention includes both the Genecept Assay and the G-DIG tool. The registry does not describe a comparison that separates the contribution of testing from the contribution of the decision tool, nor does it describe a control condition. Even if numerical changes were available, an uncontrolled change over time would not isolate which component, if any, was responsible.

The target population is also specifically defined by medication use and mental health treatment. This makes clear who the registry intended to study, but it does not guarantee that the findings would apply to people with different medication patterns or clinical circumstances. The record does not report participant characteristics beyond the eligibility description, so the representativeness of the enrolled group cannot be assessed in detail.

Statistically, a continuous primary outcome requires more than a scale range to support interpretation. Readers need the observed summaries or an estimated change, along with information about variability and uncertainty. The registry names the mean and standard deviation but reports neither value. This makes it impossible to assess the size or precision of the primary outcome from the record.

The planning material also illustrates why design assumptions need to be distinguished from the design actually used. The planning illustration assumes equal allocation between two arms, while the study is not described as a randomized, like-for-like comparison. Planning calculations do not create a control group and do not represent observed power. With outcome estimates and analysis details absent, the registry record is most informative as a description of a clinical research question and less informative as evidence about the size or reliability of any effect.

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.

Several limitations specific to this study constrain interpretation. The registry describes an open-label, naturalistic design and does not state an allocation method or describe a like-for-like control group. Participants and providers were therefore not described as blinded, and the record does not provide a controlled comparison to distinguish intervention-associated change from change over time or other aspects of care. The combined use of pharmacogenetic testing and the decision tool also makes their separate contributions impossible to assess from the design description.

The main limitation of the posted results is that numerical outcome values are not reported. The record does not give Clinical Global Impression means, standard deviations, or a change estimate at baseline and 12-weeks. It also does not report values for medication count, PHQ-9, or GAD-7. Without estimates and uncertainty measures, neither the size nor the precision of changes can be evaluated.

Further analysis details are not reported: the statistical model, analysis population, missing-data methods, multiplicity approach, and detailed outcome analysis plan. The registry also does not describe the Clinical Global Impression rating procedure or the adverse-event definitions and collection process. These gaps prevent readers from assessing how measurements were obtained, how incomplete data were handled, or how inferential choices may have affected results.

The enrollment was 53, but the record does not state the number analyzed for each outcome. It does not report participant characteristics beyond the eligibility description, limiting assessment of how well the enrolled group represents the intended population. The registry's completed status and study dates do not resolve these outcome and analysis-reporting gaps.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Medication Optimization Using Pharmacogenetic Testing and the G-DIG to Reduce Polypharmacy in a Mental Health Population — NCT03468309.
• Registry record last updated 2024-04-23; status last verified 2023-02.
• Results first posted 2022-02-28.
• 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

NCT03468309 is a completed, open-label, naturalistic study sponsored by the Seattle Institute for Biomedical and Clinical Research. The recorded start date is 2018-01-01, primary completion date is 2019-04-01, and enrollment is 53. The study evaluated the Genecept Assay with the G-DIG decision tool in veterans prescribed 5 or more medications under the stated mental health eligibility criteria.

Clinical Global Impression was the primary outcome, assessed at baseline and 12-weeks, with a stated range of 1-6. Secondary outcomes were number of psychiatric medications, PHQ-9, and GAD-7. The registry record does not report numerical results for these outcomes, a detailed statistical analysis plan, missing-data handling, or an analysis population. It lists 0 participants affected by serious adverse events, 0 affected by other adverse events, and 0 deaths among 53 at risk. The design is not described as a randomized, like-for-like comparison.