PRIME Care (PRecision Medicine In MEntal Health Care): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for PRIME Care (PRecision Medicine In MEntal Health Care) (NCT03170362), a randomized na study. The primary outcome is Depression Remission.

2. Clinical Question

The clinical question is about whether making pharmacogenetic information available when antidepressant treatment is being considered is associated with different measured depression outcomes over the trial period. It is not a comparison of two drug regimens, because the registry identifies the intervention as a pharmacogenetic test and describes its role as supporting provider choice.

The primary endpoint is a threshold-based measure of remission at the 24-week assessment. The posted comparison is between participants in the intervention group and participants in the delay results group. The registry record does not report additional detail about treatment decisions made after providers received results, such as whether a result changed medication choice or how often it was used. Such information would help distinguish the effect of access to the test from the effect of particular subsequent clinical decisions.

3. Trial Design

PRIME Care was a randomized, parallel-group trial with single masking. The registry lists the study as completed and reports enrollment of 1944 participants. The condition was major depression, and the sponsor was the VA Office of Research and Development. The registry dates the study start to 2017-06-15 and primary completion to 2021-10-30.

Participants were assigned to an intervention group or a delay results group. For the intervention group, pharmacogenetic test results were to be provided to the provider within 72 hours of randomization, with the stated purpose of facilitating antidepressant selection. The delay results group did not receive results at randomization; results were scheduled to be provided after the 24-week assessment. The registry record does not identify which people were masked under the single-masking designation.

The primary outcome was assessed at 24 weeks after randomization. Its reported analysis denominators were 754 in the intervention group and 787 in the delay results group. These are the denominators attached to the posted primary outcome results; they are distinct from the overall enrollment figure. The registry record does not report why the primary outcome denominators differ from overall enrollment, or whether this reflects missing outcome measurements, analysis-population rules, or another reason. It also does not report the trial’s original sample-size justification or detailed allocation procedure.

The registry lists depression severity and depression response as secondary outcomes, both assessed at 24 weeks. There are no posted analysis rows in the registry record for these outcomes. The results presented for the secondary endpoints are therefore descriptive group summaries and unadjusted comparisons, rather than a record of a detailed analysis plan. The registry record does not report whether other follow-up assessments or additional outcomes were planned beyond those listed.

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.

6. Primary Endpoint

The registry lists the primary outcome as Depression Remission, measured over 24 weeks post randomization.

The primary endpoint is the number of participants reaching depression remission at 24 weeks after randomization. Remission was defined using the Patient Health Questionnaire 9 (PHQ9), a self-assessed measure whose scale ranges from 0 to 27, with lower scores indicating less depression severity. A score of 5 or less at the endpoint met the remission threshold. The registry reports the number reaching that threshold in each group, along with the group denominators.

A threshold-based endpoint gives a clinically interpretable count of participants meeting a stated symptom criterion at a particular assessment. It differs from comparing average scale scores: it classifies each participant according to whether the prespecified cutoff was met, rather than retaining all score differences as a continuous measure. The remission result therefore describes the proportion meeting the threshold, not the size of each participant’s change from baseline. Baseline PHQ9 values and individual changes are not reported in the registry record.

Using a self-assessed symptom measure offers a structured way to describe depression status at follow-up, but the endpoint reflects the chosen scale, cutoff and assessment time. A score at or below the remission threshold is the trial’s operational definition; it should not be interpreted as a broader claim about every aspect of a participant’s health or functioning. The registry record does not report additional validation or endpoint-adjudication details.

7. Results Posted in the Registry

• Delay Results Group: 16.0% (126/787) versus 17.2% (130/754) on control; risk difference -1.2 percentage points (95% CI -5.0 to 2.5), risk ratio 0.93 (0.74–1.16); unadjusted two-proportion z-test p = 0.516.

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.

Depression Severity

• 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:
  • Delay Results Group: mean 4.80 versus 5.40 on control; difference -0.60 (95% CI -1.17 to -0.03), standardized effect d = -0.10; z-test p = 0.041.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for Delay Results Group than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Depression Response

• Endpoint type: Binary: the number of participants with the event, compared as proportions.
• Test: Unadjusted two-proportion z-test, with a Wald 95% confidence interval for the risk difference.
• Result:
  • Delay Results Group: 27.4% (216/787) versus 32.1% (242/754) on control; risk difference -4.6 percentage points (95% CI -9.2 to -0.1), risk ratio 0.86 (0.73–1.00); unadjusted two-proportion z-test p = 0.046.
• Interpretation:
  • The 95% confidence interval excludes no difference, so the proportion with the event is statistically lower for Delay Results Group than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

The secondary outcomes add information about depression severity and response, but they do not form a single consistent pattern that reinforces the primary remission comparison. The posted group summaries and comparisons should be read according to each endpoint’s own definition and scale: severity is a continuous score, while response and remission are threshold-based participant counts. These outcomes are related to depression but are not interchangeable measures of the same event.

In the registry results, the severity and response comparisons have unadjusted p-values near conventional significance thresholds. They may be of interest as exploratory or supportive findings, but the primary remission comparison remains the principal outcome identified for the trial. The secondary findings do not reverse or resolve the uncertainty in that primary result. The registry record does not report a prespecified hierarchy that would assign confirmatory status to either secondary outcome.

Multiple statistical comparisons create an interpretive concern: when several outcomes are examined, the chance of observing at least one apparently unusual result under a collection of null hypotheses can increase. Adjusting for multiplicity is one way to address this, while a prespecified testing hierarchy is another. No adjustment, hierarchy or other multiplicity strategy is reported in the registry record. As a result, the secondary p-values should be treated as unadjusted evidence, not as standalone confirmation of a treatment effect.

The secondary results also do not establish why a difference might occur. The registry record does not report medication choices following test disclosure, adherence information, or an analysis connecting those decisions to symptom outcomes. The secondary outcomes therefore describe measured group results, not a causal pathway from test information to treatment selection and then to depression status. Their role is to complement the primary endpoint while preserving the distinction between a registered primary question and additional outcome comparisons.

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

• Depression Severity
• Depression Response

9. Statistical Interpretation

10. Statistical Methods Explained

For a binary endpoint such as remission, each participant is classified as either meeting or not meeting the stated threshold. The group summaries report the number meeting the threshold and the denominator contributing to that result. A risk difference expresses the contrast between the two groups’ proportions. In this comparison, the posted direction is delay results group relative to intervention group. A negative risk difference therefore indicates a lower observed proportion in the delay results group; a positive value would indicate a higher observed proportion in that group. The risk difference is expressed on the probability scale, which makes it possible to interpret the estimate directly as an absolute difference between group proportions.

A confidence interval describes a range of effect estimates compatible with the observed data under the statistical model and its assumptions. It conveys precision as well as direction: a narrower interval indicates greater precision, while a wider interval leaves more uncertainty about the effect’s size. If the interval for a risk difference includes zero, the data are compatible with no group difference as well as with differences in either direction. The interval is not a guarantee that the true effect lies within its limits, and it does not describe the range of individual participant outcomes.

The registry also posts unadjusted p-values. A p-value measures how unusual data at least as discrepant as those observed would be under a specified null hypothesis and statistical model. It is not the probability that the null hypothesis is true, nor does it express clinical importance. A small p-value can occur for a small estimated effect when data are sufficiently informative, and a larger p-value does not prove that there is no effect. Here, the primary p-value is 0.516; interpretation should consider it alongside the risk difference and confidence interval, not in isolation.

For the continuous depression severity outcome, each group is summarized by a mean and standard deviation. The mean describes the central value of the scores, while the standard deviation describes their spread among participants. The mean difference is reported as delay results group minus intervention group. Because lower PHQ9 scores indicate less depression severity, a negative mean difference corresponds to a lower group mean in the delay results group. A difference in group averages does not imply that every participant had a change in that direction, and the posted means do not show the distribution of individual scores.

The response endpoint is another binary outcome, reported as the number of participants meeting the response definition. The registry record does not state the precise operational definition of depression response. The comparison can therefore be read as a contrast in the posted event proportions, but the meaning of response beyond the listed outcome title is not reported in the registry record. This is an important distinction from remission, for which the registry gives an explicit PHQ9 cutoff.

Relative measures, such as a risk ratio or odds ratio, express group contrasts on a different scale from the risk difference. A risk ratio compares the proportions directly as a ratio, while an odds ratio compares odds. These relative measures can be useful in other settings, but they do not replace the absolute difference; their values can be less intuitive for understanding how many participants met an outcome. The primary comparison also has relative measures posted, but the risk difference and its confidence interval provide a direct scale for the absolute group contrast.

The posted comparisons are described as unadjusted. An adjusted analysis could account for prespecified covariates or design features, but the registry record does not report such a primary model or its covariates. Nor does it report whether the primary analysis followed an intention-to-treat population, how missing endpoint data were handled, or whether any sensitivity analyses were planned. These omissions limit assessment of how robust the posted summary is to modeling and analysis-population choices.

For continuous outcomes, over-dispersion means that variability is greater than a particular statistical model expects. In general, failing to account for over-dispersion can make standard errors too small and confidence intervals too narrow. That concept is not a reported feature of this trial’s posted analyses: the continuous severity outcome is summarized with means and standard deviations, and no over-dispersion model is described in the registry record. Likewise, the record does not report interim analyses, stopping rules, covariate adjustment details or the original statistical analysis plan. These methods cannot be inferred from the randomized design alone.

11. Confidence Intervals and Precision

The primary risk difference is -0.012, and its 95% confidence interval runs from -0.050 to 0.025. The interval crosses zero, so the posted data are compatible with no difference as well as with a difference in either direction. Its lower limit is consistent with a greater remission proportion in the intervention group, while its upper limit is consistent with a greater proportion in the delay results group. The estimate is close to zero, but the interval is the more informative description of the range of uncertainty.

Precision is not the same as whether a result is statistically significant. The primary confidence interval communicates that the estimate does not pin down a single direction with certainty. The p-value of 0.516 is consistent with that reading, but neither statistic establishes equivalence. An equivalence conclusion would require an appropriate design and a stated margin for what counts as a sufficiently small difference; such a margin is not reported in the registry record.

The secondary severity mean difference has a confidence interval from -1.175 to -0.025. The response risk difference has a confidence interval from -0.092 to -0.001. Both intervals are close to zero at one limit, and the posted unadjusted p-values are 0.041 and 0.046. These intervals describe uncertainty for their respective outcomes, but they should be interpreted cautiously because the registry record does not report a multiplicity procedure or a hierarchy for secondary testing.

Confidence intervals apply to estimated group-level contrasts, not to individual predictions. They do not show the full range of participant scores, nor do they tell whether the endpoint was missing for participants outside the reported denominators. The registry record does not report the confidence-interval method or missing-data procedure. The posted limits are therefore best read as the reported precision for these comparisons, with the analysis details and population limitations kept in view.

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 sample-size table gives planning illustrations in participants per arm, not total events. Its calculations use a two-sided significance level of 0.05, equal allocation between two arms and a normal approximation. The assumed effect to detect is the stated fraction of the observed effect. These are planning assumptions for the illustration; the trial’s original design assumptions and sample-size justification are not reported in the registry record.

The 80% power column gives the estimated participants per arm for a design that would have an 80% probability of detecting the specified effect if that effect were the true one and the planning assumptions held. The 90% power column uses the same idea with a higher target probability of detecting that effect. Power is a design property under assumed conditions, not a measure of certainty about the trial’s observed result.

Each row plans for a different fraction of the observed effect. Planning to detect a smaller effect requires more participants per arm, because a smaller group contrast is harder to distinguish from random variation. The entries should therefore be read as hypothetical design requirements tied to the chosen effect and assumptions, rather than as a recommendation or a retrospective verdict on the trial’s design. This is not observed power.

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, p₀ and p₁ are the control and treatment proportions, and p̄ is their average.

Worked calculation for this trial

1. Control proportion p₀ = 0.172; treatment proportion p₁ = 0.160; average p̄ = 0.166.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.5265 = 1.0320.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5265 = 0.4431.
4. n = (1.0320 + 0.4431)² / (-0.0123)² = 14,353.2, rounded up to 14,354 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
p0 <- 0.172413793103448                          # control proportion
p1 <- 0.16010165184244                          # treatment proportion
n_per_arm <- function(power, frac = 1) {
  q1 <- p0 + (p1 - p0) * frac               # smaller effect = fraction of the observed difference
  pbar <- (p0 + q1) / 2
  z_b <- qnorm(power)
  ceiling(round((z_a * sqrt(2 * pbar * (1 - pbar)) + z_b * sqrt(p0 * (1 - p0) + q1 * (1 - q1)))^2 / (q1 - p0)^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 (unrounded n per arm)
cat("power.prop.test:", round(power.prop.test(p1 = p0, p2 = p1, power = 0.80)$n, 1), "\n")

# Simulation: how often does the z-test reject at the 80%-power sample size?
set.seed(3170362)
n    <- n_per_arm(0.80)
reps <- 20000
x0 <- rbinom(reps, n, p0); x1 <- rbinom(reps, n, p1)
ph <- (x0 + x1) / (2 * n)
z  <- ((x1 - x0) / n) / sqrt(ph * (1 - ph) * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(z) > z_a, na.rm = TRUE), reps))

Output from running the code

100% of observed effect: n per arm = 14,354 (80% power), 19,215 (90% power)
 75% of observed effect: n per arm = 25,707 (80% power), 34,414 (90% power)
 50% of observed effect: n per arm = 58,264 (80% power), 77,998 (90% power)
power.prop.test: 14353.2 
Simulated power at n = 14,354 per arm: 0.804 (20000 replicates)

The simulation draws many pairs of event counts at the planned sample size and counts how often the two-proportion z-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. Multiplicity, Missing Data and Analysis Populations

The registry lists a primary remission outcome and two secondary outcomes, but does not report a multiplicity adjustment or a hierarchy for testing them. When several outcomes are analyzed, interpreting each p-value as if it were the only planned test can overstate the strength of evidence. A prespecified strategy can clarify which comparison is primary and how additional tests are to be interpreted. Without a reported strategy, the secondary p-values are most appropriately treated as unadjusted findings and not as separate confirmatory conclusions.

There are no posted analysis rows in the registry record for the secondary outcomes. The registry provides summary results and unadjusted comparisons, but does not report a detailed statistical analysis plan, model specification, covariates, or whether any outcome analyses were prespecified in the protocol. Interim analyses and stopping rules are not reported in the registry record. This absence does not establish that such procedures were or were not used; it means their role cannot be evaluated from the registry entry.

The primary endpoint denominators are 754 and 787, while overall enrollment is 1944. The registry record does not explain the relationship between those totals, identify reasons for participants not appearing in the primary denominators, or state how missing endpoint observations were handled. It does not report imputation, complete-case rules, sensitivity analyses, or whether missingness was assumed to be related to measured or unmeasured outcomes. Missing data could affect the interpretation of a measured group comparison if the participants with available outcomes differ from those without them, but the registry information is insufficient to assess that possibility.

The record also does not specify the analysis population, such as whether participants were analyzed according to randomized assignment or another rule. Although the trial is listed as randomized, randomization alone does not disclose the handling of participants after allocation. Further details on analysis populations, missingness, outcome ascertainment and multiplicity would be needed to evaluate how the reported comparisons relate to the original trial plan.

15. 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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: hospitalization (1.9%), incarceration (0.1%).

The adverse-event summaries use at-risk denominators of 966 for the intervention group and 978 for the delay results group. The registry reports 76 and 70 participants affected by serious adverse events, respectively. It reports 650 and 636 participants affected by other adverse events, and 2 deaths in each group. These safety denominators differ from the denominators for the posted primary remission analysis, and the registry record does not explain the relationship between them.

The serious adverse event terms listed most often are hospitalization, at 1.9%, and incarceration, at 0.1%. These terms describe reported categories; the record does not state that they were caused by the intervention or by delayed access to results. Nor does it report the timing, duration, severity details, or participant-level overlap among event categories. Counts of affected participants should not be assumed to equal counts of distinct events unless the registry defines them that way.

These figures are descriptive safety summaries. Differences in counts alone do not establish a causal relationship or a meaningful group difference, especially without details of event definitions, ascertainment, follow-up exposure and statistical comparisons. The registry record does not report a formal comparative safety analysis or additional safety interpretation.

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

The primary result describes the proportion meeting a specific PHQ9-based remission threshold at the 24-week assessment among participants included in the posted group summaries. The estimated risk difference is close to zero, and the confidence interval includes zero. The result therefore does not establish a clear difference between earlier and delayed access to pharmacogenetic test results on this endpoint, but it also does not prove that the groups are equivalent.

The contrast concerns the timing of access to test information, not a comparison of one named antidepressant with another. The intervention was intended to support provider choice, and the delay group was scheduled to receive test results after the 24-week assessment. The posted outcomes do not show which medications were selected, how test reports affected provider decisions, or whether particular decisions were associated with later symptom scores. Those details are not reported in the registry record.

The findings are specific to the outcomes, follow-up time and analysis denominators described in the registry. Remission is defined by the study’s stated score cutoff; depression response is separately listed, but its precise definition is not reported. The mean severity score offers another view of symptoms, but group means do not describe each participant’s trajectory. These measures should not be treated as direct evidence about every aspect of functioning, quality of life, or long-term outcomes.

The results also do not establish that the test has no value in other settings, or that it improves outcomes for every participant. The registry summary does not support claims about subgroups, individual treatment choices, or outcomes beyond the reported measures. The trial’s completed status indicates that the registry lists the study as completed; it does not add information about unreported analyses or outcomes.

A careful reading separates the observed estimates from broader conclusions. The primary comparison is uncertain in direction and magnitude; the secondary comparisons are unadjusted and should be interpreted with less weight. The findings describe the posted trial results and should not be translated into medical advice or a claim that a particular treatment approach works or does not work.

17. Why This Trial Matters Statistically

This trial illustrates why a randomized comparison still needs careful attention to the endpoint and the contrast being tested. The intervention was access to pharmacogenetic test results at treatment initiation, while the comparator delayed access until after the 24-week assessment. The trial therefore addresses the timing and availability of information in treatment selection, rather than the direct effect of a single medication.

It also highlights the value of reporting absolute effects with confidence intervals. A point estimate alone can obscure uncertainty, and a p-value alone does not convey the range of plausible group differences. The primary estimate is near zero, but its interval extends to values in either direction. That combination is more informative than declaring a result simply positive or negative.

The secondary results demonstrate why outcome hierarchy and multiplicity matter. Different depression measures can show different patterns, and a collection of unadjusted p-values should not be read as if each were an independent confirmation. Transparent reporting of a prespecified analysis strategy, missing-data handling and the population included in each outcome analysis would strengthen interpretation.

Finally, the registry separates clinical outcomes from the pathway by which the intervention might affect care. Results about remission, severity and response do not show how frequently providers used test information or how it changed antidepressant selection. A statistical reading should preserve that distinction between measured endpoint contrasts and explanations for those contrasts.

18. 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 trial-specific details needed for a fuller appraisal are not reported in the registry record. These include the eligibility criteria and participant characteristics, the identity of the group or groups masked under the single-masking designation, and details of the allocation procedure. The overall enrollment is 1944, while the posted primary analysis uses denominators of 754 and 787; the record does not explain the difference or describe the primary analysis population.

The registry does not report the original sample-size justification, the trial’s statistical analysis plan, interim analyses, stopping rules, covariate adjustments, or the method used to calculate confidence intervals. It also does not report how missing endpoint data were handled, whether sensitivity analyses were conducted, or the rationale for any departures from the randomized assignment in analysis. These omissions limit assessment of potential bias and robustness.

The operational definition of depression response is not reported, and details of baseline outcome values or participant-level changes are absent from the posted results. For safety, the record provides affected-participant counts and selected serious-event terms, but does not describe timing, event overlap, attribution or comparative testing. It is therefore not possible to infer event causation from the posted counts.

The reported outcomes are measured at 24 weeks, and the registry summary does not provide information about longer-term outcomes or how treatment decisions changed after test results were disclosed. These are limits on what can be concluded from the registry results, rather than evidence that the information was or was not collected elsewhere.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: PRIME Care (PRecision Medicine In MEntal Health Care) — NCT03170362.
• Registry record last updated 2024-05-28; status last verified 2024-05.
• Results first posted 2024-05-28.
• 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.

22. Record Summary

PRIME Care was a completed, randomized, parallel-group trial of pharmacogenetic test-result access for antidepressant selection in major depression. The intervention group was scheduled to have results provided to providers within 72 hours of randomization; the delay results group was scheduled to receive results after the 24-week assessment. The registry reports enrollment of 1944 and lists the VA Office of Research and Development as sponsor.

The primary outcome was depression remission at 24 weeks, defined as a PHQ9 score of 5 or less. The posted risk difference is -0.012, with a 95% confidence interval from -0.050 to 0.025 and an unadjusted p-value of 0.516. The interval includes no difference. Secondary results concern depression severity and response, but their unadjusted comparisons should be interpreted cautiously because multiplicity procedures are not reported.

Details about the original sample-size justification, analysis populations, missing data, masking, and statistical analysis plan are not reported in the registry record. The posted results describe the measured outcomes and do not establish how providers used test information or explain the pathway to any group differences.