Retrospective Analysis of Outcomes With a Pharmacogenomic Algorithm: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Retrospective Analysis of Outcomes With a Pharmacogenomic Algorithm (NCT01632267), a registered phase not stated study. The primary outcome is Number of Outpatient Visits.

2. Clinical Question

The study title and summary place the analysis in the context of antidepressant prescribing and healthcare costs. Its stated interest is whether a reporting system based on genetic variation relates to outcomes and healthcare use in people with depression or anxiety. The registered primary endpoint, however, is outpatient visits; direct and indirect costs are discussed in the brief summary but are not listed as the primary endpoint. Medical absence days and disability claims are named as secondary outcomes.

Because the record does not describe a randomized comparison or a control group with corresponding results, the intended question cannot be read as a causal treatment comparison. The available registry description identifies a retrospective evaluation, but does not provide enough information to determine how patients were selected, how exposure to the reporting system was defined, or how potential differences between patients were addressed. Most importantly, numerical results for the named outcomes are not reported in the registry record.

3. Trial Design

This is a completed retrospective analysis titled “Retrospective Analysis of Outcomes With a Pharmacogenomic Algorithm.” The official title describes outpatients with major depressive disorder in a staff model HMO and a pharmacogenetic algorithm for psychotropic pharmacotherapeutic applications. Depression and anxiety are listed as conditions, and the described group consists of people with a primary diagnosis of depression or anxiety disorder.

The registry enrollment is 104. The brief summary separately describes a one-year retrospective study involving 96 patients with a DSM-IV-TR depressive or anxiety disorder. The record does not explain the difference between these figures, so they should not be treated as interchangeable counts or used to infer an analysis population. Further details about eligibility, recruitment, participant flow, or exclusions are not reported in the registry record.

The registry gives a start date of 2011-01 and a primary completion date of 2011-04, and marks the study completed. The primary outcome timeframe is the one-year study window from April 1, 2010 to April 1, 2011. The summary describes the project as a one year retrospective study, but does not explain the relation between that outcome window and the registry dates.

The record does not describe a randomized comparison. Allocation is not stated, and the registry explicitly cautions that the arms are not a like-for-like control comparison. One group description is listed: “Depression and anxiety.” No separate intervention and control groups with distinct outcome results are described. The sponsor is Assurex Health Inc.

Three outcome measures are recorded: the primary measure of outpatient visits and the secondary measures of medical absence days and disability claims. The registry reports zero analyses. It does not report a statistical analysis plan, model specifications, adjustment variables, or a sample-size justification. The record’s completion status does not, by itself, establish that the registered outcomes were analyzed or that results are available.

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

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 Number of Outpatient Visits, measured over During the one year study window (April 1, 2010 to April 1, 2011).

The primary endpoint is the number of outpatient healthcare visits during the one-year study window, April 1, 2010 to April 1, 2011. It is described as a continuous measure summarized by a mean and standard deviation, with units described as “units on a scale.” The endpoint is a count of visits, so its practical meaning is healthcare use over the specified period, not a direct measure of symptoms, remission, or medication response.

Outpatient visits may be used in a health-services analysis because they describe the amount of care recorded during a defined observation window. In this study’s context, the brief summary also discusses direct and indirect healthcare costs. Visit counts could be relevant to utilization, but the registry record does not say that the endpoint was converted into costs or that a cost analysis was reported.

General statistical context: A mean gives the average value of an outcome in a group, while a standard deviation describes how spread out individual values are around that average. For a count outcome, the distribution can be uneven: many people may have relatively few visits while a smaller number have substantially more. That pattern can make the mean and standard deviation an incomplete account of the distribution. The registry record does not provide a distribution, median, range, or other summary that would clarify the pattern in this study.

No mean, standard deviation, group-specific visit count, estimated contrast, confidence interval, or p-value for the primary endpoint is reported in the registry record. As a result, the magnitude and precision of outpatient use in the analyzed population cannot be described from the posted results.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does a mean describe?

The registered primary outcome is a count of outpatient healthcare visits, summarized as a mean and standard deviation. The mean describes the average count in the group being analyzed. The standard deviation describes the spread of individual counts around that average. Those quantities answer different questions: an average can describe overall utilization, while the standard deviation indicates whether individual participants’ counts vary considerably or cluster more closely.

General statistical context: A mean is sensitive to unusually large values. Visit counts may be right-skewed when a relatively small number of people have many visits. In that situation, reporting only a mean and standard deviation may not show the full shape of the data. A median, range, or distributional plot can sometimes add context, but none of those summaries is reported in the registry record for this outcome.

Why can over-dispersion matter for a count outcome?

General statistical context: Count models often begin with assumptions about how the variance relates to the mean. Over-dispersion means the observed variability is greater than an assumed model allows. If that mismatch is ignored, uncertainty estimates can be too small, making confidence intervals too narrow and statistical tests too optimistic. Alternative models or robust uncertainty estimates may be considered when justified by the data and analysis plan.

The registry record does not name a count model, report a distributional assessment, or describe a method for handling over-dispersion. It specifies that the primary measure is summarized by a mean and standard deviation, but that alone does not reveal what inferential method, if any, was used. Because the record reports zero analyses and no outcome estimates, it is not possible to assess whether the chosen method fit the observed data.

What would a confidence interval communicate?

General statistical context: A confidence interval expresses uncertainty around an estimate under the assumptions of the statistical method. Its width reflects precision: a narrower interval indicates more precise estimation under those assumptions, while a wider interval indicates less precision. The interval also helps readers assess which effect sizes remain compatible with the data. It does not establish that an effect is important, prove that there is no effect, or resolve bias in the study design.

No confidence interval for outpatient visits or for either secondary outcome is reported in the registry record. There is consequently no posted interval from which to judge the precision of an estimated association.

What can a p-value establish?

General statistical context: A p-value measures how incompatible the observed data are with a specified null hypothesis under the assumptions of the test. It is not the probability that the null hypothesis is true, and it does not measure the size or practical importance of an effect. A small p-value cannot correct a weak comparison or systematic bias; a larger p-value does not demonstrate that groups are equivalent.

No p-value or statistical test for the outcomes is reported in the registry record. The recorded count of analyses is zero. It would therefore be inappropriate to claim statistical significance, absence of a difference, or equivalence.

Why does the comparison structure matter?

General statistical context: A randomized comparison can help balance measured and unmeasured characteristics between groups, on average, when implemented as planned. A retrospective comparison does not obtain that protection simply by comparing recorded outcomes. Differences in baseline health, prior utilization, prescribing, or observation can be associated with both exposure and outcome. Statistical adjustment can address measured factors under suitable assumptions, but cannot automatically remove bias from unmeasured factors or inaccurate records.

The registry record says this is not a randomized comparison and that the arms are not a like-for-like control comparison. Allocation is not stated. No adjustment strategy, covariate list, or model is reported. These omissions prevent an assessment of whether the analysis addressed confounding or how any reported contrast should be interpreted. In this record, there is no posted contrast to evaluate in any case.

How should planning illustrations be understood?

General statistical context: A sample-size planning illustration asks how many participants might be needed under specified assumptions to detect a chosen effect with a chosen error rate and power. Such an illustration depends on the assumed effect and design. It is not an observed result and does not establish the probability that a completed study has detected an effect. The registry’s planning description says its illustrations use the observed effect, a two-sided significance level of 0.05, equal allocation, and a normal approximation; it also distinguishes per-arm planning quantities from total event counts for time-to-event outcomes.

Those assumptions are described as planning illustrations, not as the trial’s original design or a reported sample-size justification. The record does not report a sample-size justification. Since no observed effect estimate is reported, no particular planning value or implication can be interpreted from the record. Planning quantities should not be described as observed power.

In general, the statistical methods needed to interpret an outcome include a clearly defined analysis population, a model appropriate to the outcome, an estimate with a measure of uncertainty, and transparent handling of missing data and multiple outcomes. The registry record does not report those analysis details for this study. The absence of those details should not be filled by assuming a standard method was used.

10. Confidence Intervals for Each Arm

No two-arm comparison is shown for this trial, but each arm's mean can be given with a t-based 95% confidence interval computed from the posted mean, spread and number of participants.

R code and output

arm <- c("Depression and Anxiety")
n   <- c(97)
m   <- c(13)
sp  <- c(9.4)
sd  <- sp      # posted standard deviations
for (i in seq_along(n)) {
  se <- sd[i] / sqrt(n[i]); h <- qt(0.975, n[i] - 1) * se
  cat(sprintf("CARM|%s|%d|%.3f|%.3f|%.3f|%.3f\n", arm[i], n[i], m[i], sd[i], m[i] - h, m[i] + h))
}

CARM  Depression and Anxiety  97  13.000  9.400  11.105  14.895

11. Multiplicity, Missing Data and Analysis Populations

The registry records three outcome measures: outpatient visits as the primary outcome, and medical absence days and disability claims as secondary outcomes. When several outcomes are evaluated, the chance of obtaining at least one apparently notable result can increase if each is tested separately without a plan for interpreting the collection of tests. This is the general statistical issue of multiplicity.

The registry record does not report a multiplicity adjustment, a hierarchy among secondary outcomes, a prespecified testing sequence, or an analysis plan. It also reports zero analyses. Therefore, there are no posted outcome tests to assess and no basis for determining how any multiplicity was handled. The three registered outcomes should not be taken as evidence that three analyses were completed.

Missing-data methods and outcome-specific denominators are not reported in the registry record. Although enrollment is 104 and the brief summary refers to 96 patients, the record does not explain the difference, indicate which participants contributed to each outcome, or describe whether records were incomplete. No imputation approach, complete-case rule, or sensitivity analysis is described.

The analysis population is also not defined. The registry does not report whether all enrolled participants were included, whether exclusions occurred, or whether the group description corresponds to the population used for any analysis. Without participant flow and outcome-specific counts, readers cannot assess how missingness or selection may have influenced the results. These issues remain unresolved alongside the absence of posted outcome estimates.

12. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

The registry lists one group called “Depression and Anxiety.” For that entry, 104 participants are listed as at risk and zero as affected in the serious-event field; the other event field likewise lists 104 at risk and zero affected. These are the figures recorded for the two categories. The record does not provide serious-event terms, and the deaths field is not reported in the registry record.

A zero in an affected field means that the registry entry lists no affected participants for that category. It should not be expanded into a claim that no adverse events of any kind occurred, that harms were systematically ruled out, or that the reporting system is safer than an alternative. The record does not describe how adverse events were collected, the observation period for safety, or whether events were adjudicated.

The same group is listed for both event categories, rather than as distinct treatment arms. These figures therefore do not provide a comparative safety estimate. The registry record also does not report an analysis of adverse events or a confidence interval. The safety entries are descriptive registry fields and should be read narrowly, in light of the missing detail and the absence of a like-for-like comparison.

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

The record establishes that a completed retrospective study was registered to examine outcomes among outpatients with depression or anxiety in relation to a pharmacogenetic interpretive reporting system. It identifies outpatient visits during the one-year study window as the primary endpoint and lists medical absence days and disability claims as secondary outcomes. It also records a total enrollment of 104 and describes a group with a primary diagnosis of depression or anxiety disorder.

It does not establish what the observed outpatient visit counts were. No group means, standard deviations, estimates of association, confidence intervals, or p-values are reported. The registry records zero analyses. It therefore does not show whether the reporting system was associated with more or fewer visits, nor does it quantify how precise any association might be. The secondary outcomes also have no posted numerical results.

The description of the reporting system’s intended function is not a result showing that it predicted antidepressant response in this study. Nor does the record demonstrate a change in prescribing, healthcare utilization, cost, symptoms, or remission. The brief summary’s explanation of why treatment failures and healthcare costs are of interest is background, not evidence of an observed effect.

The retrospective design and absence of a like-for-like control comparison are important limits on causal interpretation. The registry does not state allocation or describe methods for controlling confounding. It also does not explain the difference between enrollment of 104 and the 96 patients named in the brief summary, or identify the analysis population. These gaps make it impossible to determine exactly who contributed data to each outcome.

The appropriate conclusion is limited: the registry record describes the study question and outcome definitions, but does not report results that answer the question. The listed serious-event fields show zero affected among 104 at risk in each recorded category, while other safety details are not reported. Neither those entries nor the study’s completed status establish a treatment benefit, lack of benefit, or comparative safety conclusion.

14. Why This Trial Matters Statistically

This study is statistically notable because it combines a retrospective healthcare-use question with a proposed pharmacogenetic reporting approach. The primary endpoint is outpatient visits, which is an observable measure of utilization over a defined period. The summary also frames interest in direct and indirect costs, while the registered secondary outcomes include medical absence days and disability claims. Together, these measures illustrate how clinical questions can extend beyond symptom scores to the use of healthcare and other recorded outcomes.

The record also illustrates why the distinction between a study objective and a study result matters. A rationale for using genetic information can motivate an evaluation, but it does not substitute for an estimated outcome, uncertainty interval, or transparent comparison. Here, the outcome definitions are recorded, but numerical results and analyses are not. Statistical interpretation must therefore stop short of evaluating the magnitude or precision of any association.

Design details are just as important as endpoint choice. The registry says the study is not randomized, allocation is not stated, and the arms are not a like-for-like control comparison. That cautions against reading any potential association as a causal effect. In a retrospective utilization study, patient characteristics and patterns of observation may influence both exposure and visit counts; the record does not describe an adjustment strategy.

The difference between the registry enrollment of 104 and the 96 patients described in the brief summary highlights the importance of participant flow and analysis populations. Without an explanation, readers cannot tell whether the counts refer to distinct stages or populations. Missing-data handling and outcome-specific denominators are also not reported.

Finally, a registered outcome count and an analysis count answer different questions. The presence of three outcome measures does not establish that results were analyzed or posted. This record is a useful example of why conclusions depend on accessible results and methods, not only on a compelling clinical rationale or a completed status.

15. 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 immediate limitation is that the registry record does not report numerical results for the primary or secondary outcomes. It records zero analyses, so there is no estimate to interpret and no measure of uncertainty. The direction, magnitude, and precision of any association between the reporting system and outpatient visits, medical absence days, or disability claims cannot be determined.

The registry enrollment of 104 differs from the 96 patients described in the brief summary. The record does not explain the discrepancy or identify the population used for each outcome. It provides no participant-flow details, outcome-specific denominators, or missing-data methods. This leaves uncertainty about completeness of records and who was included in any analysis.

The design is retrospective and not a randomized comparison. Allocation is not stated, and the registry cautions that the arms are not a like-for-like control comparison. The record does not describe a comparator group with distinct results, baseline characteristics, adjustment variables, or an approach to confounding. It is consequently not possible to assess whether differences in patient characteristics or observation contributed to any apparent association.

The primary outcome is a count of outpatient visits, summarized as a mean and standard deviation, but the record does not provide the distribution or a statistical model. It does not say whether the count was assessed for skewness or over-dispersion. The precise use of the reporting system, treatment decisions, follow-up practices, and linkage between the genetic report and recorded care are also not described in the registry record.

The record lists secondary outcomes but does not report a multiplicity strategy or a prespecified testing order. It does not provide a sample-size justification, interim-analysis plan, or detailed statistical plan. Finally, the safety entries lack event terms and collection details, and the deaths field is not reported in the registry record. These limitations restrict both clinical and statistical interpretation.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: Retrospective Analysis of Outcomes With a Pharmacogenomic Algorithm — NCT01632267.
• Registry record last updated 2016-01-05; status last verified 2013-02.
• Results first posted 2013-03-22.
• 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.

19. Record Summary

NCT01632267 is a completed retrospective study sponsored by Assurex Health Inc. It concerns outcomes in outpatients with depression or anxiety in relation to a pharmacogenetic interpretive reporting system. The registry enrollment is 104; the brief summary describes 96 patients, and the record does not explain the difference.

The primary outcome is the number of outpatient visits during April 1, 2010 to April 1, 2011, summarized as a mean and standard deviation. Medical absence days and disability claims are listed as secondary outcomes. The registry describes the study as nonrandomized, states that allocation is not stated, and cautions that the arms are not a like-for-like control comparison.

Numerical results and analyses are not reported in the registry record. The serious-event fields list zero affected among 104 at risk in each recorded category; deaths and further safety details are not reported. The record does not describe missing-data handling, an analysis population, or a statistical analysis plan.