The Pharmacist Follows You and Your Medication From Hospital to Your Daily Life and Investigate What This Means to You: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for The Pharmacist Follows You and Your Medication From Hospital to Your Daily Life and Investigate What This Means to You (NCT03079375), a randomized na study. The primary outcome is Readmissions.

2. Clinical Question

3. Trial Design

The registry describes a randomized controlled trial with three groups: usual care, basic intervention, and extended intervention. Allocation is recorded as randomized, masking as double, and the model as factorial. The record does not explain who was masked, how masking was maintained, or how the factorial label relates to the three described groups. Those details should not be inferred from the design labels alone.

The study summary says that 1500 participants were estimated for inclusion; enrollment is recorded as 1499. The posted group sizes for the primary outcome are 498 for usual care, 493 for the basic intervention, and 476 for the extended intervention. The trial is marked completed. Its recorded start is 2013-09-01, and its primary completion date is 2015-04-24.

The intervention contrast is clinically and statistically layered: the basic group received medication review, while the extended group received medication review plus a discharge interview and follow-up. Usual care is the reference group in the posted comparisons. Because no direct basic-versus-extended comparison is reported, the results do not by themselves quantify the incremental effect of adding the interview and follow-up to medication review.

The registry’s primary outcome is readmission within 30 days. The summary describes a broader aim involving readmissions and emergency department visits after discharge, including longer follow-up. The available outcome results also include some outcomes at 180 days. Detailed information on the randomization procedure, allocation concealment, setting, eligibility criteria, protocol deviations, and the trial’s original analysis plan is not reported in the registry record.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Overall Study

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 Readmissions, measured over 30 days.

The primary endpoint counts participants who had been readmitted within 30 days. It is a participant-level event count: each arm reports the number of participants with a readmission and the number in that arm. The registry does not report the number of separate readmission episodes, so the endpoint should not be interpreted as a count of all hospital stays.

Readmission is relevant to the study’s stated concern about medication management during the transition from hospital to general-practice care. A participant-level measure can summarize whether a readmission occurred during the specified follow-up period, and the group comparison can describe differences in the observed event proportions. It does not, by itself, establish why a participant was readmitted or whether a readmission was related to medication.

The registry defines the primary outcome as the number of patients readmitted and gives a timeframe of 30 days. It does not report a more detailed endpoint definition, rules for classifying readmissions, or information about how complete follow-up was assessed. The summary also refers to emergency department visits, but results for that measure are not reported in the registry record. The interpretation should therefore stay focused on the posted 30-day readmission outcome.

7. Results Posted in the Registry

• Basic Intervention: 19.9% (98/493) versus 22.3% (111/498) on control; risk difference -2.4 percentage points (95% CI -7.5 to 2.7), risk ratio 0.89 (0.70–1.14); unadjusted two-proportion z-test p = 0.352.
• Extended Intervention: 14.3% (68/476) versus 22.3% (111/498) on control; risk difference -8.0 percentage points (95% CI -12.8 to -3.2), risk ratio 0.64 (0.49–0.84); unadjusted two-proportion z-test p = 0.001.

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.

Drug-related Admissions

• 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:
  • Basic Intervention: 19.3% (95/493) versus 19.3% (96/498) on control; risk difference -0.0 percentage points (95% CI -4.9 to 4.9), risk ratio 1.00 (0.77–1.29); unadjusted two-proportion z-test p = 0.998.
  • Extended Intervention: 15.8% (75/476) versus 19.3% (96/498) on control; risk difference -3.5 percentage points (95% CI -8.3 to 1.2), risk ratio 0.82 (0.62–1.08); unadjusted two-proportion z-test p = 0.149.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Basic Intervention and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Extended Intervention and the control. The interval gives the range of effects that remain compatible with the data.

Drug-related Readmissions

• 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:
  • Basic Intervention: 6.9% (34/493) versus 7.6% (38/498) on control; risk difference -0.7 percentage points (95% CI -4.0 to 2.5), risk ratio 0.90 (0.58–1.41); unadjusted two-proportion z-test p = 0.656.
  • Extended Intervention: 5.0% (24/476) versus 7.6% (38/498) on control; risk difference -2.6 percentage points (95% CI -5.6 to 0.5), risk ratio 0.66 (0.40–1.08); unadjusted two-proportion z-test p = 0.098.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Basic Intervention and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Extended Intervention and the control. The interval gives the range of effects that remain compatible with the data.

Mortality

• 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:
  • Basic Intervention: 8.5% (42/493) versus 10.0% (50/498) on control; risk difference -1.5 percentage points (95% CI -5.1 to 2.1), risk ratio 0.85 (0.57–1.25); unadjusted two-proportion z-test p = 0.409.
  • Extended Intervention: 11.3% (54/476) versus 10.0% (50/498) on control; risk difference 1.3 percentage points (95% CI -2.6 to 5.2), risk ratio 1.13 (0.79–1.63); unadjusted two-proportion z-test p = 0.510.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Basic Intervention and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Extended Intervention and the control. The interval gives the range of effects that remain compatible with the data.

The secondary results address drug-related admissions, drug-related readmissions, and mortality, with outcomes recorded at different follow-up times. Taken together, they do not reproduce a single consistent pattern across every measure. Some estimates are close to no difference, while other point estimates are in the direction of fewer events in an intervention group. Their intervals are generally compatible with a range of effects, including no difference, so the secondary outcomes do not provide a uniformly precise extension of the primary result.

These outcomes are related to the study’s broader interest in medication problems and subsequent hospital use, but they are not interchangeable with the primary endpoint. A general readmission does not necessarily indicate a drug-related readmission, and drug-related admission is a separately defined outcome. Mortality is another distinct endpoint. The registry record does not provide further endpoint definitions or the adjudication process used to classify drug-related events.

All the posted secondary comparisons are unadjusted recalculations from the recorded counts. The record does not report a multiplicity plan or adjusted secondary analyses. When many outcomes and group comparisons are examined, chance findings become more likely if each is assessed separately without accounting for the full set. This is general statistical context, not a claim that a particular posted finding is false. The secondary results should therefore be considered less conclusive than a prespecified, fully reported analysis with a stated approach to multiple comparisons. They provide context for the primary result but do not replace it or establish which intervention component accounts for any observed pattern.

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

• Drug-related Admissions
• Drug-related Readmissions
• Percentage of Medication Changes Accepted by GPs
• Medication Review Changes Accepted by Physicians (in Hospital)
• Mortality
• Drug Related Mortality

9. Statistical Interpretation

10. Statistical Methods Explained

What does a risk difference describe?

A risk difference is the difference between the proportions of participants with an event in two groups. In these results, the comparison is intervention minus usual care. A negative estimate therefore corresponds to a lower observed event proportion in the intervention group; a positive estimate corresponds to a higher observed event proportion. The measure is expressed as a proportion, not as a count of participants or a relative change.

For the primary outcome, the posted risk-difference estimate for basic intervention is -0.024, and for extended intervention it is -0.08. These estimates are calculated from the posted arm counts and group sizes. They are unadjusted: no adjustment for participant characteristics or other covariates is represented in these comparisons. The registry record does not report an adjusted model for the primary endpoint.

What does a confidence interval add?

A confidence interval describes the uncertainty around an estimated effect under the statistical method used. It is not a range containing a fixed percentage of individual participants, and it does not guarantee that the true effect lies inside the interval. A wider interval reflects less precision than a narrower interval. Whether an interval includes no difference helps describe how compatible the estimate is with both no difference and differences in either direction, but the interval’s full range matters more than a single threshold.

The primary interval for basic intervention extends from -0.075 to 0.027, so it includes no difference. The interval for extended intervention extends from -0.128 to -0.032 and does not include no difference. These intervals describe the unadjusted risk-difference estimates. They do not account for possible multiplicity across outcomes or comparisons because no such adjustment is reported.

How should the p values be read?

A p value describes how unusual results at least as discrepant from a null hypothesis would be under that hypothesis and the statistical assumptions used. It is not the probability that the null hypothesis is true, nor is it a measure of the size or clinical importance of an effect. The registry’s comparisons give unadjusted p values of 0.352 for basic intervention and 0.001 for extended intervention. These values accompany the risk differences and confidence intervals; they should not be interpreted alone.

The registry record lists zero reported analyses. The risk differences, intervals, and p values discussed in these comparisons are unadjusted recalculations from the posted group counts, rather than a reported adjusted sponsor analysis. The record does not provide a statistical analysis plan or indicate whether any other inferential methods were used outside the posted results.

Why do event counts matter?

The endpoint is binary at the participant level: a participant is counted as having a readmission or not having one during the outcome period. The statistical comparison uses the number with an event together with the group size. Counts alone do not capture the timing or number of repeat events. The registry outcome does not report a time-to-first-event analysis or a count of recurrent admissions.

General statistical context: estimates based on event counts are more precise when the data contain more information about the event rate and group contrast. The observed group sizes and event counts determine the precision of these particular unadjusted estimates. The registry does not report missing-outcome handling, the number lost to follow-up, or whether all randomized participants were included in each endpoint analysis, so the effect of incomplete data cannot be assessed from the record.

What can and cannot be inferred from the design labels?

Randomized allocation is recorded, which identifies the intended design as a comparison among assigned groups. Double masking and factorial model are also recorded, but the registry does not describe the masked parties or the factorial structure in enough detail to explain how those features operated. The three-arm descriptions should not be extended into claims about a particular factorial analysis.

General statistical context: randomization can support comparisons of assigned groups by reducing systematic differences at allocation, while masking can limit some forms of bias. Neither label alone establishes that allocation was concealed, that masking was successful, or that all post-randomization analyses preserved the original assignment. The relevant procedures and analysis populations are not reported in the registry record.

11. Confidence Intervals and Precision

The confidence intervals show that the precision differs between the two primary comparisons. For basic intervention versus usual care, the risk-difference estimate is -0.024 and its 95% confidence interval runs from -0.075 to 0.027. The range includes no difference and spans estimates in both directions. The result is therefore not precise enough to distinguish the observed direction from a range that includes no difference under the stated unadjusted calculation.

For extended intervention versus usual care, the risk-difference estimate is -0.08 and its 95% confidence interval runs from -0.128 to -0.032. This range is entirely below no difference. It nevertheless covers a range of possible magnitudes, so the point estimate should not be treated as exact. The interval concerns the difference in event proportions; it does not express uncertainty about why readmissions occurred or whether they were medication-related.

Confidence intervals describe sampling uncertainty under the method used, not every source of uncertainty. They do not address unreported outcome loss, deviations from the protocol, or choices made in analyses that are not documented. The registry does not report an adjusted analysis or multiplicity correction, and the calculations are unadjusted. As a result, the intervals should be read as precision measures for these particular comparisons, not as a complete account of uncertainty across all outcomes and analyses.

General statistical context: more precise estimates tend to have narrower intervals, while less precise estimates tend to have wider intervals. The interval’s location and width both matter; a point estimate alone omits the range of values compatible with the data under the calculation.

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 planning figures describe illustrative future-study sizes based on the observed effect, not the original trial’s documented design assumptions. For the binary outcomes shown, the figures are participants per arm. The planning description specifies a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It considers detecting the stated fraction of the observed effect. The general sample-size description notes that time-to-event calculations give total events using the Schoenfeld method; the other calculations give participants per arm.

The 80% power column gives the illustrative participants per arm for a design with an 80% chance of detecting the effect assumed in that scenario, if the assumptions hold. The 90% power column gives the corresponding number for a design with a 90% chance under those assumptions. These are planning probabilities for a hypothetical study, not probabilities that the observed result is correct. The figures are not observed power.

Each scenario scales the assumed effect relative to the observed estimate: planning for 75% of the observed effect means targeting a smaller effect than the full observed estimate, and planning for 50% means targeting a still smaller one. Smaller effects are harder to distinguish from random variation, so they require more participants under the same planning assumptions. This is why the illustrative sample sizes increase as the target effect becomes smaller. Such calculations are sensitive to the starting estimate, which may itself be uncertain, and should be understood as illustrations rather than as a record of the trial’s original sample-size justification. That justification and the original design assumptions are not reported in the registry record.

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 = 0.223; treatment proportion p₁ = 0.199; average p̄ = 0.211.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.5769 = 1.1306.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.5766 = 0.4853.
4. n = (1.1306 + 0.4853)² / (-0.0241)² = 4,492.6, rounded up to 4,493 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 <- 0.22289156626506                          # control proportion
p1 <- 0.198782961460446                          # 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(3079375)
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 = 4,493 (80% power), 6,014 (90% power)
 75% of observed effect: n per arm = 8,071 (80% power), 10,805 (90% power)
 50% of observed effect: n per arm = 18,347 (80% power), 24,561 (90% power)
power.prop.test: 4492.6 
Simulated power at n = 4,493 per arm: 0.806 (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 ten outcome measures and notes six secondary outcomes as posted. It does not report the statistical analysis plan, the intended hierarchy among outcomes, or a procedure for controlling the chance of false-positive findings across multiple outcomes and comparisons. The primary endpoint is identified, but the record does not explain whether it was the only confirmatory endpoint or how the secondary endpoints were intended to be interpreted.

No reported analyses are listed. The numerical comparisons discussed for the posted outcomes are unadjusted recalculations from group counts. There is no reported model specification, covariate adjustment, sensitivity analysis, subgroup analysis, or multiplicity adjustment. These omissions do not establish that such analyses were never performed; they mean that they are not reported in the registry record.

Information about missing data and analysis populations is also absent. The registry does not report how many participants had outcome data available, how missing outcomes were handled, whether participants were analyzed according to randomized assignment, or how protocol deviations were addressed. The posted group sizes are available, but they do not by themselves show whether each outcome used every randomized participant or whether the analysis populations differed across endpoints.

General statistical context: examining multiple outcomes can increase the chance that at least one result appears notable by chance when each comparison is assessed separately. A prespecified analysis plan helps readers understand which findings were primary and how the full set of tests was handled. Without those details, the p values should be interpreted as unadjusted comparisons rather than as evidence that accounts for the broader set of outcomes.

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

The posted primary results describe differences in observed 30-day readmission proportions between usual care and each intervention group. The estimate for basic intervention is uncertain and its confidence interval includes no difference. The estimate for extended intervention is below no difference, and its confidence interval does not include no difference. These are unadjusted comparisons based on the registry’s posted counts, rather than results from a reported adjusted analysis.

The result does not establish that the extended intervention caused the difference, nor does it show which element of the extended intervention is responsible. The extended group received a package of medication review, a medication interview before discharge, and follow-up; the registry does not report separate tests of those components. Nor does the result directly compare the extended intervention with the basic intervention.

The endpoint is all-cause readmission as described in the primary outcome, not necessarily a medication-related readmission. The trial summary also refers to emergency department visits, but results for that outcome are not reported in the registry record. Secondary results are available for selected outcomes, but they involve separate definitions and timeframes and should not be treated as substitutes for the primary endpoint.

The estimates also do not answer questions about longer-term outcomes beyond those posted, the experience of participants, the resource implications of the interventions, or how results might vary across settings. The registry record does not provide enough information to assess missing data, the original analysis plan, or multiplicity handling. The findings are therefore best read as a description of the posted group counts and their unadjusted statistical uncertainty, not as a complete evaluation of every study objective or intervention component.

16. Why This Trial Matters Statistically

This trial raises a useful statistical question about how to interpret a multi-group intervention study when the intervention components differ in scope. Usual care is compared with medication review alone and with a more extensive package that also includes a discharge interview and follow-up. The group structure makes it important to distinguish comparisons against usual care from a direct comparison between intervention strategies. The registry reports the former, not the latter.

The primary outcome also illustrates why an effect estimate should be read alongside its confidence interval. The basic-intervention estimate is compatible with a range that includes no difference, while the extended-intervention estimate has an interval below no difference. Looking only at event counts or p values would leave out the scale and precision of the comparisons. Looking only at one estimate would also conceal uncertainty.

The registry’s lack of reported analyses, analysis-plan details, and multiplicity handling limits how strongly the posted results can be interpreted across the full outcome set. Those reporting details matter because several distinct outcomes and comparisons appear in the record. The trial is therefore a practical example of separating descriptive results from confirmatory claims, and of treating secondary findings with care when the prespecified testing strategy is not documented.

Finally, the study’s broad objective includes outcomes beyond the posted primary result. Keeping the interpretation tied to the endpoint and timeframe actually reported avoids extending a 30-day readmission comparison to emergency department use, longer-term effects, or specific medication-related mechanisms.

17. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The registry record leaves important trial details unavailable. Eligibility criteria, participant characteristics, recruitment settings, allocation concealment, the identities of masked parties, and the practical implementation of the interventions are not reported. The “factorial” model label is not explained alongside the three described groups, so the design’s structure cannot be reconstructed from that label alone. The condition classification is Cross-Sectional Study, while the study description identifies a randomized controlled trial; the record does not explain this classification.

There is no reported statistical analysis plan, adjusted analysis, or analysis population description. Missing outcome data, losses to follow-up, protocol deviations, and the handling of participants without outcome measurements are not reported in the registry record. These gaps prevent assessment of whether the posted group counts represent all randomized participants for every endpoint.

The study summary mentions emergency department visits, but such results are not reported. The available outcomes also do not provide further definitions for readmission or drug-related events, and the registry does not describe how drug-relatedness was determined. The primary result is limited to the specified 30-day timeframe; it should not be used to infer an unreported longer-term primary outcome.

Adverse-event fields list zero participants at risk and zero affected for serious and other adverse events in each group. Those entries do not provide an informative safety comparison when the recorded at-risk counts are zero. Mortality is reported separately as a secondary outcome. Overall, the registry supplies useful event counts but limited information for judging implementation, missingness, analysis choices, and the full range of outcomes.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: The Pharmacist Follows You and Your Medication From Hospital to Your Daily Life and Investigate What This Means to You — NCT03079375.
• Registry record last updated 2019-12-02; status last verified 2019-11.
• Results first posted 2019-12-02.
• 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.

21. Record Summary

NCT03079375 is recorded as a completed randomized, double-masked trial with a factorial model and three described groups: usual care, basic intervention, and extended intervention. Enrollment is 1499; the study summary states an estimated inclusion of 1500 participants. The primary endpoint is the number of participants readmitted within 30 days.

The posted primary comparison for basic intervention versus usual care has a risk difference of -0.024, with a 95% confidence interval from -0.075 to 0.027 and an unadjusted p value of 0.352. For extended intervention versus usual care, the risk difference is -0.08, with a 95% confidence interval from -0.128 to -0.032 and an unadjusted p value of 0.001. These comparisons are unadjusted recalculations from the posted counts. The registry record does not report analyses, a statistical analysis plan, or multiplicity handling.