Addressing Social Determinants of Health & Diabetes Self-Management in Vulnerable Populations: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Addressing Social Determinants of Health & Diabetes Self-Management in Vulnerable Populations (NCT03802825), a randomized na study. The primary outcome is Percent of Participants With A1C < 8%.

2. Clinical Question

3. Trial Design

This was a randomized, parallel-group study with no masking. Kaiser Permanente was the sponsor, and the condition listed was diabetes. The registry lists 110 participants enrolled. The brief summary describes a plan to randomize 100 patients, so the record contains both an enrollment figure of 110 and a planned number of 100. It does not explain the difference.

The primary comparison was patient navigation alone versus patient navigation combined with diabetes self-management support. Both interventions were described as lasting 6 months. In the combined arm, participants were also referred to Project Access NOW, which was intended to connect them with a community-based organization. The registry record does not report further detail about the delivered content, attendance, or amount of support received in each arm.

The recorded study start was 2019-03-01, primary completion was 2021-06-30, and status is completed. The registry identifies the study as a pilot and feasibility study. Its stated aims were to assess the feasibility of a larger pragmatic randomized trial and to examine preliminary effects on A1C reduction among racial and ethnic minority and low-income patients with poorly managed diabetes. The record does not report a formal sample-size justification or the trial's original statistical design assumptions.

For the primary outcome, the registry says that intention-to-treat analysis was used and that all participants were included. Secondary outcomes include emergency department visits, primary care visit no-show rate, medication refills, and mean A1C change. The record lists four secondary outcomes, but mean A1C change is not accompanied by arm-level results in the posted secondary results. The registry record does not report analyses for the secondary outcomes.

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:

Analysis population for the primary outcome, as defined in the registry: ITT was used for primary outcome analysis so all participants included.

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 Percent of Participants With A1C < 8%, measured over 6 months.

The primary endpoint was the percentage of participants with A1C below 8% at 6-month follow-up. A1C is the diabetes-related measure named in the study description; the endpoint turns that measure into a binary outcome by classifying each participant as either below the stated threshold or not. The registry describes an intention-to-treat analysis, with all participants included in the primary outcome population.

A threshold-based outcome can make the result easy to interpret as the proportion of participants who meet a stated target at follow-up. It does not, by itself, describe the size of each participant's A1C change. Two people whose A1C values fall on the same side of the threshold are counted alike for this endpoint even if their values differ. Conversely, a small change that crosses the threshold changes the classification. The registry lists mean A1C change as a secondary outcome, but arm-level results for that outcome are not reported in the posted results.

The primary endpoint therefore addresses the proportion meeting the threshold at follow-up, rather than the average degree of A1C reduction. Its comparison is a difference between two proportions, with uncertainty that should be read alongside the point estimate. The endpoint description specifies the 6-month time frame and intention-to-treat population, but does not report further detail about how missing measurements were handled.

7. Results Posted in the Registry

• Patient Navigation+Diabetes Self-Management Support: 33.9% (19/56) versus 33.3% (18/54) on control; risk difference 0.6 percentage points (95% CI -17.1 to 18.3), risk ratio 1.02 (0.60–1.72); unadjusted two-proportion z-test p = 0.947.

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.

Emergency Department (ED) 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:
  • Patient Navigation+Diabetes Self-Management Support: mean 1.07 versus 0.96 on control; difference 0.11 (95% CI -0.79 to 1.01), standardized effect d = 0.05; z-test p = 0.810.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Patient Navigation+Diabetes Self-Management Support and the control. The interval gives the range of effects that remain compatible with the data.

Primary Care Visit No-show Rate

• 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:
  • Patient Navigation+Diabetes Self-Management Support: mean 0.88 versus 0.85 on control; difference 0.03 (95% CI -0.52 to 0.58), standardized effect d = 0.02; z-test p = 0.914.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Patient Navigation+Diabetes Self-Management Support and the control. The interval gives the range of effects that remain compatible with the data.

Medication Refills

• 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:
  • Patient Navigation+Diabetes Self-Management Support: 44.4% (20/45) versus 60.0% (27/45) on control; risk difference -15.6 percentage points (95% CI -35.9 to 4.8), risk ratio 0.74 (0.49–1.11); unadjusted two-proportion z-test p = 0.140.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Patient Navigation+Diabetes Self-Management Support and the control. The interval gives the range of effects that remain compatible with the data.

The secondary outcomes cover health-care use and medication refills. Across the posted comparisons, the estimated differences are modest in relation to their uncertainty: the visit outcomes have mean-difference estimates close to zero, and the refill comparison points in the opposite direction from the visit estimates. The confidence intervals for these comparisons include zero. This broad pattern is consistent with the primary result's lack of a clearly distinguished arm difference, but secondary outcomes do not independently establish the primary endpoint's result.

These outcomes measure different aspects of care from the primary threshold-based A1C endpoint. Emergency department visits and missed primary care visits concern service use, while medication refills is a participant count. A pattern in one of these measures would not, by itself, demonstrate a corresponding change in A1C. Mean A1C change is listed among secondary outcomes, but arm-level results are not reported in the posted secondary results.

The registry lists no reported secondary analyses. The available comparison p-values are unadjusted, and the record does not report a multiplicity procedure or a prespecified hierarchy of secondary hypotheses. General statistical explanation: when several outcomes are examined, the chance of at least one apparently unusual result due to random variation can rise. This makes isolated unadjusted findings less persuasive than a result from a prespecified primary analysis, especially when the study is framed as preliminary.

The secondary findings are therefore best treated as descriptive and exploratory. They can inform questions for later research, but they carry less weight than the primary endpoint and should not be used to claim a benefit or harm across outcomes. Differences in the reported denominators also counsel care: the refill outcome uses 45 participants per group, and the registry record does not explain why that outcome has a different denominator from the primary analysis population.

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

• Emergency Department (ED) Visits
• Primary Care Visit No-show Rate
• Medication Refills
• Mean A1C Change

9. Statistical Interpretation

10. Statistical Methods Explained

What does the primary comparison estimate?

The primary outcome is binary: each participant either had A1C below 8% at 6 months or did not. For a binary outcome, a risk or proportion is the number meeting the definition divided by the number assessed in the relevant group. The registry reports the arm-level percentages, and the comparison is expressed as a risk difference for the combined-support arm relative to patient navigation alone. A risk difference is an absolute contrast between proportions, rather than a ratio of one group's proportion to the other's.

The posted primary comparison gives a risk difference of 0.006, with a confidence interval spanning negative and positive values, and an unadjusted p-value of 0.947. A value of zero for a risk difference corresponds to equal proportions in the two groups. A confidence interval that spans zero indicates that the observed data are compatible with differences in either direction under the interval's calculation. The p-value describes how unusual data at least as discrepant as those observed would be under a null model of no difference, given the analysis assumptions; it is not the probability that the null hypothesis is true.

How should the secondary comparisons be read?

Emergency department visits and primary care visit no-show rate are reported as means with standard deviations. Their comparisons are mean differences: the intervention-arm mean minus the patient-navigation mean. A mean difference is expressed in the outcome's original units, which are visits and missed visits in these results. The standard deviation describes participant-to-participant spread around a group mean; it is not the same as the uncertainty around the estimated mean or mean difference.

Medication refills is reported as a count of participants and is compared as a risk difference. This binary comparison uses the number with the outcome and the reported group denominator. The posted results give an unadjusted p-value for each comparison. No adjusted secondary analyses are reported in the registry. The comparisons should therefore be read as exploratory estimates rather than as a set of confirmatory tests.

What does intention to treat mean in this record?

The primary outcome description says intention-to-treat analysis was used and that all participants were included. In general, intention-to-treat analysis compares participants according to their randomized assignment, regardless of whether they fully received or followed the assigned intervention. This preserves the randomized comparison as a basis for estimating the effect of assignment. The registry record does not specify how missing primary outcome measurements were handled, or whether every randomized participant had an observed 6-month measurement.

The primary arm counts are 54 and 56. The posted medication-refill outcome uses 45 participants in each arm, while the primary outcome is described as including all participants. The record does not explain the different denominators. That difference matters when interpreting comparisons: the number contributing to an outcome may affect both the estimate and its precision.

Why do dispersion and over-dispersion matter?

General statistical explanation: count outcomes such as visits can have a distribution that is more variable than a simple model assumes. When variation is greater than expected under the model, standard errors can be too small if that extra variation is ignored, making confidence intervals too narrow and p-values too optimistic. The registry reports means and standard deviations for the visit outcomes, but it does not report a model-based analysis of those counts or an adjustment for over-dispersion. The posted mean-difference comparisons should not be mistaken for evidence that a particular count model was used.

For a binary outcome, precision depends not only on the observed proportions but also on the number of participants contributing data. For a mean difference, precision depends on the group sizes and the variability of the outcome. In general, more variability makes it harder to estimate a mean difference precisely. These principles explain why a small study can produce wide uncertainty even when the point estimate is close to zero.

What are the limits of the posted statistical plan?

The registry record does not report the original sample-size justification, the prespecified statistical analysis plan, interim analyses, or multiplicity procedures. It also does not report adjusted models for the posted outcomes. The comparison estimates and p-values available for the outcomes are unadjusted. General statistical explanation: adjustment can account for prespecified covariates or design features, while multiplicity procedures address the increased chance of chance findings when many outcomes or comparisons are examined. Without the relevant plan and analyses in the registry record, it is not possible to determine whether such procedures were part of the trial's original conduct.

The 6-month endpoint is a follow-up measurement, not a time-to-event outcome. The registry does not report event-time analyses for these outcomes. The posted comparisons should be interpreted as arm-level summaries over the specified follow-up period, not as estimates of a time-to-event effect.

11. Confidence Intervals and Precision

Precision is about how narrowly an estimate is determined by the observed data. The primary risk-difference estimate is 0.006. Its confidence interval extends across zero, so the data are compatible with a difference favoring either group as well as with little or no difference. The interval is not narrow enough to settle the direction or size of the contrast. The p-value of 0.947 is consistent with the same cautious reading, but it does not replace the interval: the interval conveys a range of values compatible with the analysis, while the p-value summarizes evidence against a null model.

Confidence intervals are not guarantees that the true effect lies within the particular interval, nor do they assign probabilities to the competing explanations after the data have been observed. General statistical explanation: under repeated use of the same method and its assumptions, intervals would cover the target quantity at the method's stated coverage rate. The registry record does not state the confidence level or the precise interval method, so neither should be inferred from the posted limits.

For the secondary comparisons, the intervals also span zero. That means the estimates do not precisely establish whether the arm differences are positive or negative. The visit outcomes have substantial reported variability relative to their small observed mean differences, while the medication-refill outcome has fewer participants contributing than the primary outcome population described in the registry. Those features are relevant to the uncertainty, but the registry does not provide missing-data explanations or adjusted estimates that would clarify their influence.

Precision should not be confused with whether a result is statistically significant. A broad interval can contain both negligible and potentially important differences. Because the record does not report a threshold for a clinically important difference, the clinical importance of the range cannot be determined from the statistical comparison alone.

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 figures are planning illustrations based on the observed effect, not the trial's original design assumptions. For the listed outcomes, figures are participants per arm, not total participants or total events. The record notes that time-to-event calculations would instead give total events, but the outcomes are not time-to-event outcomes.

The planning assumptions are a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. The effect to detect is set to the stated fraction of the observed effect. The 80% power column describes the number of participants per arm that the illustration estimates would be needed for an 80% chance of detecting an effect of that size under those assumptions, if the effect were real. The 90% power column uses a higher target chance of detection and therefore calls for more participants under the same assumptions.

Rows that plan for 100% of the observed effect use the observed estimate as the target effect; rows for 75% or 50% use smaller target effects. Smaller effects are harder to distinguish from random variation and require more participants. The very large figures for some outcomes reflect the small observed effects in this trial and should not be read as a recommendation or a feasibility judgment. These calculations are not observed power: they are prospective illustrations based on specified assumptions, not a measure of what the completed trial achieved.

The registry record does not report whether these illustrations match the sample-size rationale used when the trial was designed. The trial's own original design assumptions and sample-size justification 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 = 0.333; treatment proportion p₁ = 0.339; average p̄ = 0.336.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6681 = 1.3095.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6681 = 0.5623.
4. n = (1.3095 + 0.5623)² / (0.0060)² = 98,890.8, rounded up to 98,891 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 <- 0.333333333333333                          # control proportion
p1 <- 0.339285714285714                          # 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(3802825)
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 = 98,891 (80% power), 132,387 (90% power)
 75% of observed effect: n per arm = 175,615 (80% power), 235,098 (90% power)
 50% of observed effect: n per arm = 394,700 (80% power), 528,390 (90% power)
power.prop.test: 98890.8 
Simulated power at n = 98,891 per arm: 0.801 (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 record lists one primary endpoint and several secondary outcomes, but does not report a multiplicity plan. It does not say whether the secondary outcomes were prespecified as hypothesis tests, whether their order was fixed, or whether any adjustment was planned for examining multiple outcomes. The record also lists no reported analyses for the secondary outcomes. The unadjusted comparisons can be described as estimates, but their p-values should not be treated as if they came from a single isolated test in a study with only one question.

General statistical explanation: testing multiple outcomes creates more opportunities for a chance finding. A multiplicity strategy can control how strongly to interpret such findings, for example by specifying which outcome is primary or by adjusting testing thresholds. Without a reported strategy, the secondary p-values do not establish confirmatory evidence. This does not invalidate the descriptive estimates; it limits the strength of claims based on them.

The registry describes the primary analysis as intention to treat and says all participants were included. It does not report how missing outcome data were handled, whether measurements were available for every randomized participant, or whether any sensitivity analyses assessed the influence of missingness. For medication refills, the reported denominator is 45 participants per arm; the record does not explain why it differs from the primary outcome population. No missing-data mechanism should be assumed from the difference alone.

The registry record does not report interim analyses, a complete statistical analysis plan, adjustment covariates, or other model details. It also does not provide an analysis population description for the secondary outcomes beyond their posted arm denominators. These omissions make it difficult to distinguish planned analyses from additional exploratory analyses or to assess whether missing data and multiple testing were addressed in the original analysis. Accordingly, the most defensible interpretation is limited to the posted estimates and their stated outcome definitions, with secondary results treated cautiously.

15. Safety Results

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

Deaths

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: hospitalizations (11.1%), hypoglycemia (1.9%).

The registry reports 7 serious adverse events among 54 participants at risk in the patient-navigation group and 7 among 56 at risk in the combined-support group. It reports 0 other adverse events in each group and 1 death in each group. The listed serious-event terms include Hospitalizations at 11.1% and Hypoglycemia at 1.9%. The registry record does not specify the denominator for those term percentages in the information.

These counts describe events recorded in each group; they do not show that the interventions caused the events. The counts are similar in number, but the study is small and the registry does not report a comparative statistical analysis of adverse events. A count of deaths or serious events is not, by itself, an estimate of causal risk, and the available record does not provide enough detail to determine relatedness, timing, severity by participant, or whether an individual had more than one event.

The safety information should therefore be read descriptively and in the context of the limited follow-up and study size. It does not support a conclusion about comparative safety.

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

The result shows that the recorded percentage meeting the A1C threshold at 6 months was similar in the two randomized groups: 33.0% with patient navigation alone and 34.0% with patient navigation plus diabetes self-management support. The risk-difference estimate is 0.006, and its confidence interval crosses zero. The posted comparison therefore does not precisely identify a difference between the groups.

This is not evidence that the approaches are equivalent. A failure to distinguish an effect from zero can arise when the true difference is small, when the estimate is uncertain, or both. Equivalence requires a defined margin and an analysis designed to assess whether the uncertainty lies within that margin. Neither an equivalence margin nor a noninferiority framework is reported in the registry record.

Nor does the primary outcome establish the size of A1C changes among participants. It classifies participants according to whether their A1C was below the stated threshold at follow-up. The registry lists mean A1C change as a secondary outcome, but does not report its arm-level results in the posted secondary results. It is therefore not possible to infer the distribution of individual changes from the primary endpoint.

The study's stated purpose was preliminary: assess feasibility for a larger pragmatic randomized trial and examine preliminary effectiveness. The posted results do not report a formal feasibility assessment or the original sample-size justification. The study should not be taken as definitive evidence about results in other settings or populations. Its randomized design provides a comparison of the assigned groups, but the unmasked design, limited sample, follow-up period, and uncertainty all matter when interpreting the estimate.

The secondary comparisons and adverse-event counts add descriptive information, not a basis for a treatment recommendation. The registry record does not establish causal effects on those outcomes or report adjusted analyses. Overall, the defensible conclusion is that the observed primary outcome proportions were close, while the uncertainty remains broad enough that meaningful differences in either direction cannot be ruled out.

17. Why This Trial Matters Statistically

This trial is statistically informative because it makes the distinction between an observed estimate and a precise conclusion visible. The primary percentages are close, but a point estimate alone cannot determine whether the underlying difference is negligible or whether the study has simply estimated it imprecisely. The interval crossing zero and the small p-value evidence against a difference do not prove that the groups are equivalent; they indicate that the data do not clearly resolve the comparison.

The study also illustrates why a pilot and feasibility study should not be read as a definitive effectiveness trial. Its stated purpose included assessing the feasibility of a larger pragmatic randomized trial and examining preliminary effectiveness. Those goals are different from demonstrating a final clinical effect. The record does not report a formal feasibility result or the trial's original sample-size justification, so conclusions about those aims must remain limited.

The secondary outcomes emphasize the importance of prespecification and multiplicity. Multiple outcomes can be useful for developing hypotheses and understanding a study, but unadjusted comparisons have less confirmatory weight when no multiplicity plan is reported. The registry also contains differences in outcome denominators, without an explanation of missingness. That makes clear reporting of analysis populations and missing-data handling important to interpreting an estimate's precision.

Finally, the posted results provide an example of how safety counts and efficacy outcomes answer different questions. Event counts can describe what occurred during follow-up, but without further analysis they do not establish causation or comparative safety. Taken together, the record supports careful interpretation of randomized estimates, uncertainty, outcome definitions, and the limits of preliminary evidence.

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 features specific to this study limit how far the results can be taken. The registry frames it as a pilot and feasibility study and describes a target population of patients with poorly managed diabetes who were African-American, Hispanic, and/or Medicaid. Results in this selected population and setting may not generalize to other groups or care systems. The study was conducted in KPNW, and the record does not report results by race, ethnicity, or Medicaid status.

The study was unmasked, so participants and study personnel were not blinded according to the registry. The primary endpoint is an A1C threshold at 6 months; it does not describe longer-term outcomes or the size of A1C changes. The record lists mean A1C change as a secondary outcome but does not report arm-level results for it. A single follow-up time point also does not show whether any difference persisted.

The registry records 110 enrolled participants, while the brief summary describes a plan to randomize 100. The record does not explain this difference. The primary analysis is described as intention to treat, but the handling of missing outcome data is not reported. The medication-refill outcome has 45 participants per arm, and the reason for that denominator is not reported in the registry record.

The registry does not report the original sample-size justification, detailed analysis plan, interim analyses, or multiplicity procedures. The posted secondary comparisons are unadjusted, and the record does not report adjusted analyses. These limits affect interpretation of both precision and the evidential weight of secondary findings. The adverse-event information is also descriptive: the registry does not provide event relatedness or a comparative analysis.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Addressing Social Determinants of Health & Diabetes Self-Management in Vulnerable Populations — NCT03802825.
• Registry record last updated 2022-11-29; status last verified 2022-11.
• Results first posted 2022-11-29.
• 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

NCT03802825 is a completed, randomized, parallel-group, unmasked pilot and feasibility study sponsored by Kaiser Permanente. It enrolled 110 participants and compared patient navigation alone with patient navigation plus diabetes self-management support. The primary endpoint was the percentage with A1C below 8% at 6 months, analyzed using intention to treat according to the registry description.

The registry reports primary outcome percentages of 33.0% and 34.0%, with a risk-difference estimate of 0.006 and an unadjusted p-value of 0.947. The confidence interval spans zero, so the comparison is not precise enough to establish the direction of a difference. Secondary results and adverse-event counts are descriptive and should be interpreted cautiously. The record does not report the trial's original sample-size justification, missing-data handling, or a multiplicity plan.