PictureRx: Improving Medication Safety in Health Disparity Populations: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for PictureRx: Improving Medication Safety in Health Disparity Populations (NCT02289469), a non-randomized na study. The primary outcome is Number of Participants With Changes/Updates in Medication List.

2. Clinical Question

3. Trial Design

The registry record describes PictureRx as a parallel, nonrandomized study with no masking. It lists 488 participants enrolled and a completed status. The study start is recorded as 2015-04, and primary completion is recorded as 2016-03. The intervention is the PictureRx medication history platform; the comparator is the usual medication history process.

Nonrandomized allocation is important when interpreting the arm comparison. Without random assignment, factors associated with receiving one process rather than the other may also be related to medication-history updates. Differences in observed outcomes, if reported, would therefore not by themselves show that the platform caused those differences. The registry record does not report the allocation procedure, participant flow by arm, baseline characteristics, or methods used to adjust for differences between groups.

The background describes emergency departments as a setting where accurate medication lists can inform care and where patients may leave without an updated list. It also explains the rationale for an illustrated, patient-centered medication list and the platform’s intended availability in English or Spanish. These details establish the motivation for the research, but the registry record does not report enough trial-specific setting or participant information to determine how closely the enrolled population reflects that background.

The study is listed with phase NA. The record identifies PictureRx, LLC as sponsor. It does not report the number or location of study sites, eligibility criteria, the method used to assign participants, or whether the intervention was delivered consistently across participants. Those omissions limit assessment of how the comparison was conducted and how broadly its results might apply. Most importantly, the primary outcome results are not reported in the registry record, so the design description cannot be paired with an observed estimate of the intervention-comparator contrast.

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.

Before assignment: A total of 250 patients were enrolled into the intervention group. However, only 244 patients could be matched to suitable controls, and 6 were therefore excluded from the analysis of results. Thus the number of patients recruited was 494 (250 intervention + 244 control), but results are presented for 488 (244 intervention + 244 control).

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 Number of Participants With Changes/Updates in Medication List, measured over 1 day.

The primary endpoint is titled “Number of Participants With Changes/Updates in Medication List.” It counts participants for whom nurses made updates or corrections to the medication history in the electronic health record. The recorded timeframe is 1 day. This is a participant count, not a measure of the number of corrections per person, the size or accuracy of each correction, or whether a medication-related problem was prevented.

This type of outcome can capture whether the medication history was changed during the specified period. It is relevant to the study’s stated concern that medication lists may be incomplete or inaccurate. A count alone, however, does not describe the clinical importance of each update or establish that the resulting list was accurate. Interpretation also depends on clear and consistent definitions of an update, the assessment process, and the population included in the count. Those operational details are not reported in the registry record.

The registry record does not report the number of participants meeting this endpoint in either arm, or an estimated contrast between arms. It also does not report an uncertainty interval or a test result for the primary outcome. As a result, the endpoint’s definition can be described, but the posted record does not permit an assessment of its observed frequency or the precision of a comparison.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The study is recorded as nonrandomized, parallel, and unmasked. These features describe how the comparison was structured, but they do not supply a statistical analysis plan. The registry record does not report the specific model or test used for the primary outcome, how covariates were selected, or how the analysis handled incomplete observations. Statistical methods that might be used for this type of outcome can be explained generally, but they should not be mistaken for methods documented for this study.

What does a participant count represent?

The primary outcome is a count of participants for whom nurses made changes or corrections to the medication history in the electronic health record. In statistical terms, this can be treated as a binary outcome for each participant: the participant either met the endpoint or did not. Group-level reporting might express the outcome as counts, proportions, or a contrast between groups, but the registry record does not report which presentation or analysis was used. No unreported result should be inferred from the endpoint label.

A participant-level binary outcome differs from a count of multiple updates. A person with several corrections would still contribute as one participant if the outcome is defined as the number of participants with any qualifying change. The record’s endpoint title and description point to participants, not a tally of individual corrections. It does not specify the precise operational rule for deciding which changes qualify or how inconsistent records were handled.

Why does nonrandom allocation matter?

Random allocation can help balance both measured and unmeasured characteristics across study arms, on average. In a nonrandomized comparison, that protection is absent. For example, if participants entering one arm differed in clinical circumstances or medication-history complexity, those differences could influence the chance of an update. This is a general concern, not a claim that such differences occurred in this study. Baseline characteristics and adjustment procedures are not reported in the registry record.

Adjustment methods can account for measured characteristics when estimating an association, but they cannot automatically remove bias from unmeasured factors or poorly measured variables. The registry record does not identify any adjusted or unadjusted primary estimate. Accordingly, it is not possible to assess whether the reported analysis, if any, accounted for differences between arms.

How should uncertainty be interpreted?

A point estimate is a single summary of an observed contrast. A confidence interval conveys uncertainty around that estimate under the chosen statistical method and its assumptions. A p-value, when reported, addresses how compatible the data are with a specified null model; it does not measure the size or clinical importance of an effect. Neither a primary estimate, a confidence interval, nor a p-value for the primary outcome is reported in the registry record. The study’s enrollment and design alone cannot establish whether an estimate would be precise.

For count outcomes, over-dispersion is a general issue when the variability in counts exceeds what a simple model assumes. It can make standard errors too small if the model does not reflect that extra variability. The primary outcome is described as a count of participants meeting an endpoint, rather than the number of corrections. The registry record does not report a model, a dispersion assessment, or evidence that over-dispersion was relevant; it would be inappropriate to assert that any particular count model was used.

What do the planning assumptions mean?

The planning illustration is described as using a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It considers an effect defined as a stated fraction of the observed effect. These are planning assumptions, not evidence of the study’s achieved precision. A sample-size calculation depends on its assumed effect and analysis conditions. It cannot substitute for observed outcome counts, and it should not be called observed power. The registry record does not report a statistical plan that would establish whether these planning assumptions matched the trial’s original design.

For a binary outcome, planning calculations also depend on the assumed event frequency in each group and on the effect being targeted. Those inputs affect the number of participants estimated to be needed. The registry record does not provide enough information to evaluate the underlying outcome frequencies or confirm the assumed effect used for any numerical planning estimate. The planning assumptions are therefore best read as illustrative rather than as a description of the completed trial’s analysis.

What cannot be recovered from the record?

The registry record does not report the primary outcome results or the analysis details needed to reconstruct them. It does not specify the model, the handling of missing observations, the set of covariates, or the method for producing uncertainty intervals. Recalculation from summary statistics is not possible without the relevant arm-level outcome data. General statistical explanations can clarify what such analyses mean, but they cannot fill in trial-specific methods or results that are not reported.

10. Exact Estimates for Each Arm

No two-arm comparison is shown for this trial, but each arm's result can still be summarized with exact interval estimates computed from the posted counts. The Clopper-Pearson interval is exact and conservative; the Wilson interval is a common closer-to-nominal alternative; the Bayesian credible interval uses a uniform Beta(1, 1) prior. For small groups the three can differ noticeably.

R code and output

arm    <- c("PictureRx", "Usual Care")
n      <- c(244, 244)
events <- c(101, 43)
for (i in seq_along(n)) {
  ci <- binom.test(events[i], n[i])$conf.int          # Clopper-Pearson exact interval
  cr <- qbeta(c(0.025, 0.975), 1 + events[i], 1 + n[i] - events[i])   # Bayesian, uniform Beta(1,1) prior
  wi <- (function(x, n) { p <- x / n; z <- qnorm(0.975); d <- 1 + z^2 / n
    c((p + z^2 / (2 * n) - z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d, (p + z^2 / (2 * n) + z * sqrt(p * (1 - p) / n + z^2 / (4 * n^2))) / d) })(events[i], n[i])
  cat(sprintf("ARM|%s|%d|%d|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f|%.1f\n", arm[i], n[i], events[i], 100 * events[i] / n[i],
              100 * ci[1], 100 * ci[2], 100 * wi[1], 100 * wi[2], 100 * cr[1], 100 * cr[2]))
}

ARM  PictureRx  244  101  41.4  35.1  47.9  35.4  47.7  35.4  47.7
ARM  Usual Care  244  43  17.6  13.1  23.0  13.4  22.9  13.4  22.9

11. Multiplicity, Missing Data and Analysis Populations

The registry record lists two outcome measures and two analyses. The primary outcome concerns participants with changes or updates in the medication list; a secondary outcome is patient understanding. Listing multiple outcomes raises a general multiplicity issue: when several outcomes or analyses are examined, the chance of at least one apparently notable finding can increase if no adjustment or prespecified interpretation strategy is used. The registry record does not report whether any multiplicity adjustment, hierarchy, or other strategy was planned or applied.

The record does not report how missing primary or secondary outcome data were handled. It does not give the number of participants with missing data, reasons for missingness, imputation procedures, or sensitivity analyses. Missingness can affect interpretation if the participants whose outcomes are unavailable differ from those with observed outcomes. Without the relevant details, the direction or extent of any such influence cannot be assessed.

The registry record also does not define the primary analysis population, such as whether analysis was based on all enrolled participants or a subset with outcome data. Although enrollment is listed as 488 and the safety entries list 244 participants at risk in each arm, those figures do not establish the denominator used for the primary endpoint. The analysis populations and participant flow are not reported in the registry record. These omissions prevent assessment of whether the outcome analysis preserved the original comparison or how exclusions and missing observations may have affected it.

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 adverse-event entries list 244 participants at risk in the PictureRx group and 244 in the Usual Care group. Each group has zero recorded serious adverse events. The entries also list zero other adverse events in each group. These are the event counts recorded for the groups; they do not establish that no event occurred outside the recorded categories or observation process.

The registry record does not report details such as the safety observation period, event definitions, collection procedures, or whether the groups were assessed in the same way. Deaths are listed as null rather than as a count of zero, so death information is not reported in the registry record. The safety entries are distinct from the primary medication-list outcome and do not indicate whether nurses made changes or corrections. A zero recorded event count should be read as a registry value, not as proof of absence of risk.

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

The record establishes that a completed, parallel, nonrandomized study compared a PictureRx medication history platform with the usual medication history process. It identifies a primary outcome focused on whether nurses updated or corrected medication histories in the electronic health record within a timeframe of 1 day. It also describes the rationale for improving medication-list clarity and safety in populations facing health disparities.

What the record does not establish is the primary result. The arm-level counts for participants meeting the endpoint, the contrast between arms, and its uncertainty are not reported in the registry record. Without those data, the outcome cannot show whether the groups differed, how large any difference was, or how precisely it was estimated. The record does not support a claim that the platform improved or failed to improve the endpoint.

The nonrandomized design further limits causal interpretation. Any observed difference, if reported elsewhere, could reflect differences in the groups, the intervention, or other factors. The registry record does not provide the baseline information or analysis details needed to evaluate those possibilities. The endpoint itself also captures whether an update occurred, not whether every medication entry was correct, whether participants understood their lists, or whether patient outcomes changed.

The zero serious and other adverse-event entries are safety counts, not evidence about the primary outcome. They should be interpreted with the limits of the record’s event definitions and observation details in mind. Overall, the registry record describes the study question and design but does not report enough outcome information to determine what the comparison found.

14. Why This Trial Matters Statistically

This study addresses a statistical problem that is common in evaluations of health-information tools: a plausible process improvement is not the same as a demonstrated outcome difference. The primary endpoint records whether nurses changed or corrected medication histories. That is a concrete process measure, but a count of participants with updates does not by itself establish the accuracy of the final medication list or its effect on patient understanding or safety.

The nonrandomized design makes the comparison especially dependent on clear reporting of participant selection, baseline characteristics, and analysis methods. Without these details, readers cannot determine whether observed arm differences might also reflect differences in who received each process. Transparent reporting of outcome counts and uncertainty would be needed to assess both the magnitude and precision of any association.

The record also illustrates why an endpoint definition and a completed status are not substitutes for posted results. Enrollment is listed as 488, but the registry record does not report primary outcome counts or an effect estimate. The available safety counts answer a different question and cannot fill that gap. Similarly, planning illustrations based on assumptions are not evidence of achieved precision or a substitute for the observed comparison.

For biostatisticians and clinical researchers, the central issue is thus the distinction between study rationale, study design, and analyzed evidence. The rationale explains why medication-history support was considered; the design describes a nonrandomized comparison; the outcome record needed to evaluate the primary statistical question is not reported. Keeping these elements separate avoids overstating what the registry record can establish.

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 important trial-specific limitation in the registry record is the absence of reported primary outcome results. No arm-level counts, comparative estimate, confidence interval, or test result are reported for the endpoint. The primary question therefore cannot be answered from the record. The listed enrollment does not resolve this limitation, because the number contributing to the primary analysis is not reported.

The study is nonrandomized and unmasked. The registry record does not explain how participants were allocated, describe baseline characteristics, or report adjustment methods. This prevents assessment of potential group differences and their influence on an observed comparison. The record also does not report eligibility criteria, participant flow, study locations, or detailed implementation procedures.

The endpoint is a participant count for medication-history updates or corrections, but operational details are limited. The registry record does not specify how an update was defined, how accuracy was assessed, or how multiple changes for one participant were treated beyond describing the outcome in terms of participants. It also does not report missing-data handling or the primary analysis population. These omissions constrain reproducibility and make it difficult to judge the endpoint’s completeness.

Safety reporting is limited to the listed event counts and denominators. The observation period and event-collection methods are not reported, and deaths are not reported as a numeric count. The record identifies patient understanding as a secondary outcome but does not report its results. Together, these gaps mean the registry record supports a description of the study question and design, but not a full evaluation of effectiveness, precision, or safety.

16. Related Tutorials

17. Related Statistical Calculators

18. Sources

• ClinicalTrials.gov: PictureRx: Improving Medication Safety in Health Disparity Populations — NCT02289469.
• Registry record last updated 2017-07-07; status last verified 2017-07.
• Results first posted 2017-07-07.
• 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

The registry record describes PictureRx: Improving Medication Safety in Health Disparity Populations as a completed, nonrandomized, unmasked, parallel study with 488 participants enrolled. The study start is listed as 2015-04 and primary completion as 2016-03. PictureRx is compared with the usual medication history process.

The primary endpoint counts participants for whom nurses made updates or corrections to medication histories in the electronic health record within 1 day. Patient understanding is listed as a secondary outcome. The record does not report the primary outcome results, a comparative estimate, or the analysis details needed to assess its precision. Safety entries list zero serious and zero other adverse events in each group, with 244 participants at risk per group; death counts are not reported in the registry record.