Effect of Vyvanse on Driving in Young Adults With ADHD: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Effect of Vyvanse on Driving in Young Adults With ADHD (NCT00801229), a randomized phase 4 study. The primary outcome is Participants Experiencing Collisions During "Surprise Events" in Driving Simulator.

2. Clinical Question

3. Trial Design

This was a phase 4, randomized, double-blind, parallel-group, placebo-controlled study. Massachusetts General Hospital is listed as the sponsor. The registry lists the study as completed, with a start date of 2008-12 and primary completion in 2010-07. The recorded enrollment is 75.

The registry describes an intended study of young drivers with ADHD and a driving simulation designed to assess driving behavior, including speed, collision risk, and visual attention. The primary outcome record focuses on collisions during surprise events. The endpoint description says participants completed an initial one-hour driving simulation and a second simulator session following a 6 weeks trial on lisdexamfetamine or placebo. The registry lists Vyvanse and placebo as the intervention groups, and describes daily doses of 30, 50, or 70mg for each.

The primary outcome population was participants who completed the protocol, including the endpoint driving simulation assessment. The registry reports that 61 participants were analyzed for this outcome. The enrollment record, intended population description, and endpoint analysis population describe different aspects of the study; the registry does not explain their relationship in further detail. The trial's own original sample-size assumptions and justification are 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.

Signed Consent

Reasons for not completing this period:

Received Study Drug/Placebo

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Participants who completed the protocol, including endpoint Driving Simulation assessment, were analyzed (61 in total).

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 Participants Experiencing Collisions During "Surprise Events" in Driving Simulator, measured over 6 weeks.

The primary endpoint counts participants who experienced a collision during a surprise event in the driving simulator. These events were designed to test attention and driving performance. The endpoint is binary at the participant level: a participant either experienced a collision during the specified events or did not. The registry reports the number of participants with the event, rather than the total number of collision events.

This outcome was used to assess one aspect of simulated driving performance under a sudden attentional demand. A participant-level collision outcome provides a direct comparison of how many people in each group experienced the specified event. It does not describe the severity or number of collisions per participant, and those details are not reported in the registry record.

The outcome description refers to the difference in the number of collisions experienced by individuals in the groups, while the posted measure is participants experiencing collisions. The posted group results are 8.0 participants in the Vyvanse arm and 17.0 in the placebo arm. The analysis population is described as participants who completed the protocol and endpoint simulation assessment. The result therefore pertains to the recorded simulator endpoint and analysis population, rather than all enrolled participants or ordinary road driving.

7. Results Posted in the Registry

• Vyvanse: 25.8% (8/31) versus 56.7% (17/30) on control; risk difference -30.9 percentage points (95% CI -54.3 to -7.4), risk ratio 0.46 (0.23–0.89); unadjusted two-proportion z-test p = 0.014.

8. Statistical Interpretation

9. Statistical Methods Explained

What does a binary endpoint represent?

General statistical explanation: a binary endpoint records whether a defined event occurred for each participant. For the primary outcome, the event is experiencing a collision during a surprise event in the simulator. The posted counts describe participants with at least one qualifying event as defined by the endpoint, rather than a rate of collisions per unit of driving time. The registry does not provide additional detail about how repeated collisions within a participant were handled.

What does the risk difference communicate?

General statistical explanation: a risk difference compares the event probabilities in two groups by subtracting one group's probability from the other's. It is expressed on the probability scale and can describe an absolute contrast. The posted risk-difference estimate is negative for Vyvanse relative to placebo, consistent with the lower event count reported for the Vyvanse group. Its confidence interval describes uncertainty around that estimated contrast. The exact interval bounds are not repeated in this explanation.

Why report ratio measures as well?

General statistical explanation: a risk ratio compares event probabilities as a ratio, while an odds ratio compares event odds. Both express relative rather than absolute contrasts, and their numerical values do not have the same interpretation as a risk difference. The posted results include intervals for these measures as well as for the risk difference. Because the endpoint is binary and the observed group counts are finite, each measure provides a different scale for describing the same group comparison. They should not be treated as independent pieces of evidence or as separate clinical outcomes.

What does the p-value tell a reader?

General statistical explanation: a p-value measures how unusual data at least as discrepant as those observed would be under a specified null model and its assumptions. It is not the probability that the null hypothesis is true, and it does not give the probability that the result will replicate. The registry reports an unadjusted p-value of 0.014. The registry record does not identify the statistical test used to obtain it, describe an adjustment for multiplicity, or report a complete statistical analysis plan. The p-value should therefore be read as the posted unadjusted result, without inferring unreported methods.

Why is the analysis population relevant?

The primary outcome description states that participants who completed the protocol, including the endpoint driving simulation assessment, were analyzed; the record reports 61 participants in this population. This is a completer-based description. General statistical explanation: restricting an analysis to participants with completed endpoint assessments can affect interpretation if completion is related to group assignment or outcome. The registry does not report reasons for noncompletion, the outcome status of participants without an endpoint assessment, or an imputation or other missing-data method. It is not possible to determine from the record how those issues might affect the comparison.

What can be inferred about the model?

The trial design is described as randomized, double-blind, and parallel-group. Those labels describe allocation and masking features, not the statistical model used for the primary comparison. The specific model, any covariate adjustment, and the analysis method are not reported in the registry record. The reported comparison should not be described as adjusted or as following a particular model unless that information is documented.

General statistical explanation: randomization is intended to support a fair comparison of groups by allocating participants without choosing the group based on their characteristics. Masking is intended to limit the influence of group knowledge on participants or study personnel. These design features do not guarantee that all sources of bias are absent, and the registry description alone does not establish how allocation concealment or masking success was assessed.

10. Confidence Intervals and Precision

The posted results include confidence intervals for the risk difference and ratio measures. A confidence interval expresses the statistical uncertainty around an estimate under the method used to calculate it. A narrower interval indicates greater precision on that measure's scale; a wider interval indicates less precision. An interval is not a guarantee that the true effect lies within its bounds, and its interpretation depends on the analysis assumptions.

The risk-difference interval extends from a negative value to a negative value, so the interval as posted does not include zero. The ratio-measure intervals are also reported. Their interpretation depends on the relevant null value for a ratio, rather than the null value for a difference. The registry record does not name the interval calculation method, so no specific method should be inferred.

The observed group counts are 8.0 participants with collisions in the Vyvanse arm and 17.0 in the placebo arm. Although the posted unadjusted p-value is 0.014, the p-value alone does not convey the precision of the effect estimate or the range of plausible effect sizes. The confidence intervals provide that complementary information. Precision is also specific to the endpoint, analysis population, and statistical method; it does not indicate how well the result generalizes to other driving outcomes or settings.

11. 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 are participants per arm, not total events. They are planning illustrations based on the observed effect, not the trial's original sample-size plan. The listed 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 specified as a fraction of the observed effect.

For planning at 100% of the observed effect, the illustration gives 39 participants per arm for 80% power and 52 participants per arm for 90% power. For planning at 75% of the observed effect, it gives 72 participants per arm for 80% power and 96 participants per arm for 90% power. For planning at 50% of the observed effect, it gives 164 participants per arm for 80% power and 219 participants per arm for 90% power.

In plain terms, the 80% column gives the planned participants per arm under the stated assumptions for a design with 80% power to detect the specified effect. The 90% column gives the corresponding planning count for 90% power. Power is the chance, under the assumed effect and statistical model, of detecting an effect at the specified significance level. Smaller effects require more participants to detect with the same power under these assumptions. These figures are not observed power; they are forward-looking planning illustrations, and the trial's own original design assumptions are not reported in the registry record.

12. 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 = 0.567; treatment proportion p₁ = 0.258; average p̄ = 0.412.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.6962 = 1.3645.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.6611 = 0.5564.
4. n = (1.3645 + 0.5564)² / (-0.3086)² = 38.7, rounded up to 39 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 <- 0.566666666666667                          # control proportion
p1 <- 0.258064516129032                          # 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(801229)
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 = 39 (80% power), 52 (90% power)
 75% of observed effect: n per arm = 72 (80% power), 96 (90% power)
 50% of observed effect: n per arm = 164 (80% power), 219 (90% power)
power.prop.test: 38.7 
Simulated power at n = 39 per arm: 0.808 (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.

13. Multiplicity, Missing Data and Analysis Populations

The registry lists no secondary outcomes posted and no analyses posted. It does not report whether multiple outcome measures or analyses were planned, whether the primary endpoint was selected before the data were examined, or whether a multiplicity adjustment was considered. The reported p-value is explicitly unadjusted. Without the planned testing strategy, the possibility and handling of multiple comparisons cannot be assessed from the registry record.

The primary endpoint population is described as participants who completed the protocol, including the endpoint driving simulation assessment. The record reports 61 participants in that analysis population. The enrollment entry gives 75, but the registry does not explain the relationship between enrollment and endpoint completion. Reasons for missing endpoint assessments, group-specific missing-data counts, and methods for handling missing data are not reported in the registry record.

The record does not report whether the analysis followed an intention-to-treat approach, whether participants were analyzed according to randomized assignment, or whether any other analysis population was used. It also does not report an interim-analysis plan, stopping rules, or the statistical analysis plan. The trial's original sample-size justification and assumptions are not reported in the registry record. These omissions limit assessment of how the posted analysis relates to the study's planned design and whether decisions about outcomes or analyses could have affected the reported comparison.

General statistical explanation: multiplicity matters because testing several outcomes or comparisons can increase the chance of observing a result that appears statistically notable by chance. Missing-data methods matter because the interpretation of an analysis can depend on why data are absent and how the analysis addresses them. These are general considerations; the registry record does not establish that any specific problem occurred in this study.

14. Safety Results

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

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 3% reporting threshold.

Most frequent non-serious adverse event terms

The registry reports serious adverse events for 35 participants at risk in the Vyvanse group and 34 at risk in the placebo group, with 0 affected in each group. This is a count of participants reported as affected, not a description of event types or event rates. The record lists no serious adverse event terms.

For other adverse events, the record reports 30 affected among 35 at risk in the Vyvanse group and 20 affected among 34 at risk in the placebo group. These are group-level counts; the registry excerpt does not define the individual adverse-event terms or provide further detail about severity, timing, or attribution. The serious-event and other-event fields are separate categories in the record and should not be combined into a new total.

Death counts are not reported in the registry record. The posted safety counts describe what is recorded in this registry entry, and they do not alone establish a complete safety profile or a causal relationship between an intervention and an event.

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

The registry result documents fewer participants with a collision during surprise events in the Vyvanse group than in the placebo group: 8.0 compared with 17.0. The posted unadjusted p-value is 0.014, and the comparison includes confidence intervals for the effect measures. These statistics describe the observed comparison and its uncertainty for the stated simulator endpoint and analysis population.

The endpoint is limited to collision experience during surprise events in a driving simulator after the trial period described as 6 weeks. It does not directly measure routine driving on public roads, crashes outside the simulator, or long-term driving outcomes. Although the study description also discusses speed, lane maintenance, traffic-signal behavior, and visual attention, results for those outcomes are not posted as secondary outcomes in the registry record.

The analysis population consists of participants who completed the protocol and endpoint driving simulation assessment. The record does not report missing-data handling or the outcome status of participants who did not complete that assessment. The registry's enrollment entry and the endpoint analysis population should therefore not be treated as interchangeable.

The result does not show that the observed group contrast will occur in every setting or population, and the posted comparison alone cannot establish causality beyond the limits of the trial design and analysis. It also does not establish that Vyvanse improves real-world driving or that it has no effect on other driving measures. Those broader conclusions are not supported by the posted primary endpoint alone. The findings are best read as a comparison of the recorded simulator outcome in the analyzed study population, with statistical uncertainty and reporting limitations taken into account.

16. Why This Trial Matters Statistically

This trial illustrates how a clinical question about a complex behavior can be translated into a defined statistical endpoint. Driving performance includes many components, but the posted primary outcome focuses on whether a participant experienced a collision during surprise events in a simulator. That clarity makes a group comparison possible, while also placing limits on what the endpoint can represent.

The result also highlights the distinction between an absolute and a relative effect measure. The posted comparison includes a risk difference and ratio measures, each expressing the contrast on a different scale. Reading the estimate alongside its confidence interval is more informative about uncertainty than relying on the p-value alone. The p-value is unadjusted, and the registry does not report the broader analysis plan or any multiplicity strategy.

Another statistical issue is the analysis population. The primary outcome is reported for protocol completers who completed the endpoint simulation assessment. Without information about missing assessments or their handling, readers cannot evaluate how closely that analysis reflects the full randomized population. Finally, the planning illustrations show how required participant counts change when the effect to detect is smaller, under stated assumptions. They are not the trial's original design calculations and do not describe the study's observed power.

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 primary endpoint is a simulator-based outcome, specifically collisions during surprise events. It does not directly establish performance during everyday driving. The registry describes other driving behaviors as study objectives, but does not post secondary outcome results for them.

The primary analysis population is described as protocol completers with an endpoint simulation assessment, while the registry also lists enrollment of 75. The relationship between those figures is not explained in the registry record. Reasons for missing assessments and methods for handling missing data are not reported. This limits assessment of whether the completer analysis could differ from an analysis of all randomized participants.

The registry does not report the primary statistical test or model, covariate adjustment, multiplicity plan, interim analyses, or the original sample-size justification. It also does not explain the role of all possible dose levels in the analysis. The registry describes daily doses of 30, 50, or 70mg in both groups, but does not give further detail about dose assignment or dose-specific outcomes.

The safety entry reports affected-participant counts but not detailed adverse-event terms beyond the absence of listed serious-event terms. Death counts are not reported in the registry record. These limitations are about the information available in the registry entry; they do not establish that any specific methodological or safety problem occurred.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Effect of Vyvanse on Driving in Young Adults With ADHD — NCT00801229.
• Registry record last updated 2014-03-10; status last verified 2014-02.
• Results first posted 2011-08-25.
• 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

NCT00801229 is a completed phase 4 randomized, double-blind, parallel-group, placebo-controlled study sponsored by Massachusetts General Hospital. The registry lists an enrollment of 75 and study dates from 2008-12 to primary completion in 2010-07. The study examined Vyvanse versus placebo in young adults with ADHD, using a driving simulator.

The primary outcome was participants experiencing collisions during surprise events. Among protocol completers with an endpoint simulation assessment, the registry reports 8.0 participants with collisions in the Vyvanse group and 17.0 in the placebo group, with an unadjusted p-value of 0.014. The registry does not post secondary outcomes or analyses. Its planning illustrations are based on the observed effect and stated assumptions, rather than the trial's original design plan.