Targeted Adherence Intervention to Reach Glycemic Control With Insulin Therapy for Patients With Diabetes: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Targeted Adherence Intervention to Reach Glycemic Control With Insulin Therapy for Patients With Diabetes (NCT02846779), a randomized na study. The primary outcome is Insulin Persistence.

2. Clinical Question

3. Trial Design

This was a randomized, parallel-group study comparing low-, moderate-, and high-intensity pharmacist outreach. The registry record lists the study as completed and gives an enrollment of 6000. The sponsor was Brigham and Women's Hospital. The study start is recorded as 2016-07 and primary completion as 2017-12-30. The phase is listed as NA.

The intervention descriptions indicate increasing follow-up intensity, but the details are not complete. Low intensity involved quarterly educational mailings and limited pharmacist telephone outreach. Moderate intensity added more frequent pharmacist follow-up, optional text messaging, and limited contact with the participant's provider. High intensity added still more pharmacist follow-up and more follow-up with the provider and/or pharmacist. The descriptions also refer to targeting a subset of participants in the moderate- and high-intensity groups, but the record text is truncated. The full targeting procedures are not reported in the registry record.

Masking is listed as none. The study summary says the strategies were equivalently priced and varied in targeting and intervention intensity. The registry record does not report the original sample-size justification, allocation ratio, randomization procedure, concealment details, or a complete statistical analysis plan. It also does not report whether analyses followed intention-to-treat or another prespecified population definition.

The primary outcome was insulin persistence, expressed as the percentage of participants classified as non-persistent from 1 month (30 days) after randomization through 12 months (365 days) after randomization. Secondary outcomes listed include change in HbA1c, health care spending, physician office visits, emergency room visits, and hospitalizations. The study design supports comparisons among the outreach strategies; it does not by itself establish that differences in outcomes were caused solely by outreach intensity or targeting.

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 Insulin Persistence, measured over From 1 month (30 days) after randomization through 12 months (365 days) after randomization.

The primary endpoint is titled Insulin Persistence and is reported as the percentage of participants. Its operational definition focuses on non-persistence: a participant was classified as non-persistent if insulin was not refilled before a set threshold of time during follow-up. Thus, the posted percentages describe the share meeting that non-persistence definition; they are not the share who were persistent.

The outcome window began 1 month (30 days) after randomization and continued through 12 months (365 days) after randomization. The exact refill-gap threshold is not reported in the registry record. Without that detail, readers cannot assess how the persistence classification would change under a different threshold or reproduce the endpoint definition fully.

Refill-based persistence is useful for evaluating an adherence-focused intervention because it captures whether medication refills occur within a specified interval. It is an indirect measure of medication-taking: a refill record does not establish whether insulin was taken as directed, and the endpoint does not directly measure glycemic control. That distinction matters because an intervention could affect refill patterns without producing an equivalent change in HbA1c, or vice versa.

For interpretation, lower percentages of non-persistence correspond to fewer participants meeting the endpoint's non-persistence definition. The posted group percentages are 5.4% for low intensity, 4.7% for moderate intensity, and 4.9% for high intensity. They are descriptive group results; the comparative uncertainty and sponsor-reported risk ratios provide additional context about the evidence for differences.

7. Results Posted in the Registry

• Moderate Intensity: 4.7% (88/1873) versus 5.4% (100/1861) on control; risk difference -0.7 percentage points (95% CI -2.1 to 0.7), risk ratio 0.87 (0.66–1.16); unadjusted two-proportion z-test p = 0.346.
• High Intensity: 4.9% (91/1862) versus 5.4% (100/1861) on control; risk difference -0.5 percentage points (95% CI -1.9 to 0.9), risk ratio 0.91 (0.69–1.20); unadjusted two-proportion z-test p = 0.501.

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.

Change in Glycated Hemoglobin Level (HbA1c)

• 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:
  • Moderate Intensity: mean -0.21 versus -0.06 on control; difference -0.15 (95% CI -0.30 to 0.00), standardized effect d = -0.11; z-test p = 0.058.
  • High Intensity: mean -0.31 versus -0.06 on control; difference -0.25 (95% CI -0.41 to -0.09), standardized effect d = -0.17; z-test p = 0.002.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Moderate Intensity and the control. The interval gives the range of effects that remain compatible with the data.
  • The 95% confidence interval excludes no difference, so the mean is statistically lower for High Intensity than for the control. This is an unadjusted, secondary comparison, so it is suggestive rather than confirmatory.

Health Care Spending

• 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:
  • Moderate Intensity: mean 23,284 versus 22,616 on control; difference 668.00 (95% CI -2,486 to 3,822), standardized effect d = 0.01; z-test p = 0.678.
  • High Intensity: mean 22,545 versus 22,616 on control; difference -71.00 (95% CI -3,085 to 2,943), standardized effect d = -0.00; z-test p = 0.963.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Moderate Intensity 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 High Intensity and the control. The interval gives the range of effects that remain compatible with the data.

Number of Physician Office Visits

• 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:
  • Moderate Intensity: mean 7.62 versus 7.66 on control; difference -0.04 (95% CI -0.50 to 0.42), standardized effect d = -0.01; z-test p = 0.866.
  • High Intensity: mean 7.50 versus 7.66 on control; difference -0.16 (95% CI -0.63 to 0.31), standardized effect d = -0.02; z-test p = 0.503.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Moderate Intensity 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 High Intensity and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes extend beyond refill-based persistence to HbA1c, health care spending, physician office visits, emergency room visits, and hospitalizations. They address distinct aspects of glycemic control and health care use, rather than repeating the primary endpoint. The posted numerical summaries include HbA1c, spending, and office visits; endpoint-specific values and comparisons are stated with those results rather than repeated.

Across these secondary measures, the pattern is not uniform. The HbA1c summaries show larger decreases in the moderate- and high-intensity groups than in low intensity, while spending and office-visit summaries are similar across groups. This means the secondary measures do not provide one consistent account of intervention differences. They are also measured on different scales, so their estimates should not be compared by size alone.

These findings do not replace the primary persistence result. HbA1c is a clinical measure, whereas persistence is a refill-based classification; health care use outcomes capture yet other dimensions. Differences in one outcome do not establish a corresponding difference in another. The registry record does not report a unified analysis that combines these outcomes or explains how they should be prioritized relative to the primary endpoint.

Unadjusted secondary results carry less weight because multiple outcomes and comparisons create more opportunities for apparently notable findings to arise by chance. The registry record does not report a multiplicity adjustment or a prespecified hierarchy for interpreting secondary results. Accordingly, the secondary pattern is best treated as exploratory context alongside the primary comparison, not as a definitive confirmation or contradiction of it. The listed emergency-room and hospitalization outcomes should not be assigned a direction or effect from the registry record's outcome list alone.

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

• Change in Glycated Hemoglobin Level (HbA1c)
• Health Care Spending
• Number of Physician Office Visits
• Number of Emergency Room Visits
• Number of Hospitalizations

9. Statistical Interpretation

10. Statistical Methods Explained

What does the risk ratio compare?

A risk ratio compares the probability of an event in one group with the probability in a reference group. In these primary comparisons, the event is classification as non-persistent, and low intensity is the reference. A value of 1 corresponds to the same event risk in the groups; a value below 1 indicates a lower observed risk in the group named first relative to the reference, while a value above 1 indicates a higher observed risk. This is a general explanation of the statistic, not a claim about the trial's analysis procedure.

The sponsor reported risk ratios of 0.88 for moderate versus low intensity and 0.91 for high versus low intensity. The associated 95% confidence intervals are 0.75 to 1.03 and 0.77 to 1.06. These are the sponsor's posted results. The record does not identify the statistical method used to obtain them, so the method should not be inferred from the estimates alone.

What does a confidence interval convey?

A confidence interval describes uncertainty around an estimated contrast under the assumptions of the analysis. A narrower interval indicates greater precision, while a wider interval leaves more plausible values compatible with the data and method. For a risk ratio, an interval that includes 1 does not exclude equal risks. Here, both sponsor-reported intervals include 1, so the posted estimates do not rule out no difference between either more intensive strategy and low intensity.

A confidence interval is not a guarantee that the true effect lies within its limits, nor does it give the probability that a particular hypothesis is true. Its interpretation depends on the sampling and model assumptions used. Because the registry record does not report the sponsor's method, the method-specific assumptions are not reported in the registry record.

What are the recalculated comparisons?

The recalculated comparisons are unadjusted comparisons based on posted group summaries. They yield p-values of 0.346 for moderate versus low intensity and 0.501 for high versus low intensity. A p-value measures how unusual results at least as incompatible with a specified null model would be if that model and its assumptions held. It is not the probability that the null hypothesis is true, and it does not measure clinical importance.

The recalculation is not a substitute for the sponsor's analysis. It does not account for any adjustments or modeling choices that may have been used in the sponsor analysis, and the record does not report those details. The sponsor's reported risk ratios and confidence intervals remain the posted sponsor results; the recalculations are an additional, simple reading of the summary data.

Why do the outcome scale and definition matter?

The primary outcome is binary at the participant level: each participant is classified as non-persistent or not according to the refill threshold. A percentage therefore expresses how common the classification was in each group. The exact threshold is not reported in the registry record, limiting independent evaluation of the endpoint. The risk ratio compares these group risks; it does not express a change in HbA1c or a direct measure of insulin-taking behavior.

General statistical practice also distinguishes an effect estimate from its precision. A point estimate provides a central summary of the observed contrast, while its interval shows the uncertainty around that estimate. Similar-looking percentages can still have uncertain comparative estimates, and a small p-value would not by itself indicate a large or important effect. In this trial, the primary sponsor-reported intervals include the equal-risk value.

What is not known about the analysis?

The registry record does not report the prespecified model, covariate adjustment, handling of missing outcome data, treatment of participants who changed groups or stopped follow-up, or analysis population. It also does not report interim analyses, multiplicity procedures, or the original design assumptions. These omissions prevent a full audit of how the sponsor's estimates were produced and how all planned comparisons were handled. No particular approach should be assumed.

11. Confidence Intervals and Precision

The sponsor-reported 95% confidence interval for moderate versus low intensity runs from 0.75 to 1.03 for the risk ratio. For high versus low intensity, it runs from 0.77 to 1.06. In both comparisons, the interval includes 1, the value for equal risk. This means the posted estimates are compatible with no difference as well as with some relative difference in either direction under the sponsor's analysis.

The point estimates, 0.88 and 0.91, fall below 1, but the intervals are not narrow enough to establish that the true relative risk is below 1. The registry record does not provide sponsor p-values for these comparisons. The recalculated p-values are 0.346 and 0.501, and the recalculated intervals likewise include no difference. These recalculations should be distinguished from the sponsor's posted analysis.

Precision concerns the range of effects that remain compatible with the data and analysis, not whether an estimate is clinically meaningful. The primary endpoint percentages are relatively close in the posted group summaries, but a comparison should be judged using its uncertainty as well as its point estimate. The interval endpoints also do not describe all uncertainty arising from choices such as the refill threshold or missing-data handling; those details are not reported in the registry record.

The confidence intervals therefore support a cautious interpretation: the observed primary percentages are lower in the moderate- and high-intensity groups than in the low-intensity group, but the comparative estimates remain uncertain. The record does not establish a precise population effect or resolve whether intensity, targeting, or other components explain the numerical pattern.

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 illustrations are planning calculations based on the observed effects, not the trial's original design assumptions. They use a two-sided significance level of 0.05, equal allocation between the two compared groups, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect. For these non-time-to-event outcomes, the figures are participants per arm; they are not total event counts.

The 80% power column gives the estimated number of participants per arm for a future study to have an 80% chance of detecting the specified effect under the listed assumptions, if that effect and assumptions are correct. The 90% power column gives the corresponding number for a 90% chance. These are design-planning quantities, not guarantees about what a study will find. They are not observed power.

Rows describe planning for 100%, 75%, or 50% of the observed effect. Smaller target effects require more participants per arm to achieve the same planned power, because smaller differences are harder to distinguish from sampling variation. The illustrations should be read as conditional examples rather than recommendations: they rely on the observed effect as a starting point and do not incorporate design features not specified in the calculation assumptions.

The registry record does not report the trial's original sample-size justification or its original power assumptions. Therefore these illustrations cannot establish whether the actual enrollment was chosen to detect the primary contrast, whether the planned power was achieved, or whether the original design addressed multiple comparisons. They are separate from the study's observed group results and should not be interpreted as evidence for or against an intervention effect.

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 = 0.054; treatment proportion p₁ = 0.047; average p̄ = 0.050.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.3093 = 0.6062.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.3092 = 0.2603.
4. n = (0.6062 + 0.2603)² / (-0.0068)² = 16,470.0, rounded up to 16,471 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 <- 0.0537345513164965                          # control proportion
p1 <- 0.0469834490122798                          # 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(2846779)
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 = 16,471 (80% power), 22,049 (90% power)
 75% of observed effect: n per arm = 29,746 (80% power), 39,820 (90% power)
 50% of observed effect: n per arm = 67,971 (80% power), 90,993 (90% power)
power.prop.test: 16470 
Simulated power at n = 16,471 per arm: 0.803 (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. Analyses Reported by the Sponsor

The sponsor's posted primary analyses compare each more intensive strategy with low intensity using risk ratios. The reported estimate is 0.88 (95% CI 0.75 to 1.03) for moderate intensity and 0.91 (95% CI 0.77 to 1.06) for high intensity. The registry record does not give p-values for these primary comparisons or specify the methods used to obtain the estimates.

The recalculations use the posted group counts and percentages to make simple, unadjusted comparisons. Their p-values are 0.346 and 0.501, respectively, and their intervals include no difference. They provide a useful independent check at the summary-statistic level, but they do not reproduce an undocumented sponsor model or account for adjustments that may have been made. The numerical direction of the estimates is broadly consistent: both sponsor risk ratios are below 1, and the observed non-persistence percentages are lower in the more intensive groups. Both approaches also leave substantial uncertainty around the comparison.

There is no reported discrepancy between the sponsor analysis and the recalculation. Where the approaches differ in detail, the sponsor's own analysis takes precedence as the trial's reported analysis. The registry record does not provide enough methodological detail to explain how its risk ratios and confidence intervals were calculated, so additional assumptions about that analysis would be unwarranted.

15. Multiplicity, Missing Data and Analysis Populations

The study includes a primary endpoint and several secondary outcomes, with comparisons of moderate and high intensity against low intensity. Considering several outcomes and contrasts raises a multiplicity issue: even if no true differences exist, a larger collection of tests makes it more likely that some results appear notable by chance. A prespecified testing hierarchy or adjustment can help control that concern. Such procedures are general statistical approaches; the registry record does not report whether one was used.

The record does not report an adjustment for multiple comparisons, a sequence for testing the outcomes, or a clear distinction between confirmatory and exploratory analyses. The secondary results should therefore be interpreted cautiously, especially where unadjusted comparisons are discussed. The registry record also does not report interim analyses or stopping rules.

Missing-data handling is another important unknown. The primary endpoint depends on refill records over a defined follow-up window, but the record does not describe completeness of those records, censoring, exclusions, or how participants with incomplete follow-up were treated. It does not report imputation procedures or sensitivity analyses. These details matter because incomplete outcome information can affect group estimates and their uncertainty.

The analysis population is also not specified. Although allocation is randomized, the registry record does not state whether the posted analyses included all randomized participants, only participants with outcome data, or another defined group. It does not report how adherence to assigned outreach, crossover, or withdrawal was handled. Randomization supports fair comparison when preserved in analysis, but the relevant analysis choices cannot be assessed without these details.

These are reporting gaps, not evidence that a particular procedure was omitted. The registry record does not provide enough information to determine the direction or size of any resulting bias. They limit confidence in a full methodological appraisal and reinforce the value of distinguishing the posted estimates from assumptions that are not documented.

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

The primary result describes the proportion of participants classified as non-persistent under a refill-based definition during the stated follow-up window. The percentages were 5.4% for low intensity, 4.7% for moderate intensity, and 4.9% for high intensity. The sponsor's risk-ratio estimates were below 1 for both comparisons against low intensity, but both 95% confidence intervals included 1. The recalculated p-values do not provide clear evidence of a difference.

This does not show that the interventions had no effect. An interval including no difference means the results do not rule out equal risks; it does not prove equality. Conversely, the lower observed percentages in the moderate- and high-intensity groups do not establish that those strategies reduce non-persistence in the wider population. The size and direction of the population contrast remain uncertain in the posted primary analyses.

The result is specific to the endpoint definition, which uses refill timing and a threshold that is not reported in the registry record. It should not be interpreted as a direct measure of whether participants took insulin as prescribed. It also does not establish an effect on HbA1c, spending, visits, emergency care, hospitalization, or any other outcome without considering the corresponding outcome data and their uncertainty.

The study compares packages of outreach that varied in more than one way, including intensity, optional messaging, provider contact, and targeting. The results therefore do not isolate the effect of any single component. The record's intervention descriptions are incomplete, and the original analysis plan and missing-data procedures are not reported.

Overall, the posted evidence describes modest numerical differences in non-persistence, with uncertainty that includes no difference. It is not a basis for medical advice or a claim that one strategy works or does not work. Interpretation should remain limited to the trial's reported groups, endpoint, and available analysis details.

17. Why This Trial Matters Statistically

This trial is statistically instructive because it contrasts a binary adherence-related endpoint with several secondary outcomes on different scales. The primary endpoint is a participant-level non-persistence classification, while HbA1c is continuous and health care use measures describe utilization. Each requires an outcome-appropriate summary, and a pattern in one measure need not be repeated in another.

The primary group percentages are close, and the sponsor-reported relative estimates have confidence intervals that include equal risk. This illustrates why a point estimate and a statistical conclusion are not interchangeable: the observed direction can be clear in the sample while the precision is insufficient to settle the population comparison. The recalculated analyses add transparency but cannot substitute for a fully reported sponsor method.

The study also highlights the interpretive challenge of multicomponent interventions. More intensive strategies differed in follow-up, communication options, provider contact, and targeting. A comparison among those packages does not identify which element, if any, accounts for outcome differences. Better reporting of the intervention and analysis populations would aid reproducibility and help readers distinguish intervention effects from implementation differences.

Finally, the registry record demonstrates the importance of reporting endpoint definitions and statistical plans. The refill threshold, missing-data handling, multiplicity approach, and original sample-size justification are not reported. Those omissions limit interpretation even when group estimates are available. Clearer reporting would make it easier to assess precision, robustness, and the role of secondary findings without treating exploratory patterns as conclusive.

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.
• Participant counts were reconstructed from posted percentages and arm sizes, so small rounding differences are possible.

Several limitations are specific to the information recorded for this trial. The primary outcome's refill threshold is not reported, so the non-persistence definition cannot be fully evaluated or reproduced. Refill-based persistence is also an indirect measure of medication use; the registry record does not establish whether participants took insulin as directed.

The intervention descriptions are incomplete. In particular, text describing the targeting of participants in the moderate- and high-intensity groups is truncated. As a result, the exact components and reach of each strategy cannot be reconstructed from the registry record. Because intensity, communication options, provider contact, and targeting varied together, the comparisons do not isolate a single intervention component.

The sponsor's statistical method is not reported, nor are the primary analysis population, missing-data procedures, or handling of incomplete follow-up. The registry record does not provide an original sample-size justification, a multiplicity plan, or interim-analysis details. These omissions prevent a complete assessment of the analysis and of how much confidence to place in secondary comparisons.

The sponsor-reported primary confidence intervals include equal risk, and the recalculated comparisons are unadjusted. Secondary outcomes include several different measures, with no reported multiplicity procedure. The posted record also does not provide enough information to determine whether all outcome data were available for all randomized participants. These limitations constrain interpretation of the posted evidence; they do not establish that bias occurred or determine its direction.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Targeted Adherence Intervention to Reach Glycemic Control With Insulin Therapy for Patients With Diabetes — NCT02846779.
• Registry record last updated 2019-08-07; status last verified 2019-08.
• Results first posted 2019-08-07.
• 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.

22. Record Summary

NCT02846779, titled Targeted Adherence Intervention to Reach Glycemic Control With Insulin Therapy for Patients With Diabetes (TARGIT-DM), is recorded as a completed randomized, parallel-group study with no masking. Brigham and Women's Hospital is listed as sponsor. The study start is recorded as 2016-07, primary completion as 2017-12-30, and enrollment as 6000.

The study compared low-, moderate-, and high-intensity pharmacist outreach. Its primary outcome was insulin non-persistence, defined through refill timing during follow-up from 1 month (30 days) after randomization through 12 months (365 days) after randomization. The posted group percentages were 5.4%, 4.7%, and 4.9%, respectively. Sponsor-reported risk-ratio intervals included 1 for both comparisons with low intensity.

The record lists secondary outcomes covering HbA1c and health care use. Important details, including the exact refill threshold, sponsor analysis method, missing-data handling, analysis population, multiplicity plan, and original sample-size assumptions, are not reported in the registry record.