Improving Function in Age-Related Macular Degeneration: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Improving Function in Age-Related Macular Degeneration (NCT00572039), a randomized phase 3 study. The primary outcome is Targeted Vision Function (TVF).

2. Clinical Question

3. Trial Design

This was a randomized, controlled, phase 3 trial with a parallel-group design and single masking. The registry lists enrollment of 241 and a completed status. The recorded start was 2005-08, and primary completion was 2013-06. The sponsor was Thomas Jefferson University.

The trial compared two behavioral treatments for people with age-related macular degeneration: Problem-Solving Treatment and Supportive Therapy. The brief description says 120 participants were randomized to PST and 120 to ST. The total enrollment recorded in the registry is 241, so the record does not fully explain how that enrollment count relates to the stated arm numbers. The primary outcome population is described separately: 106 PST and 112 ST participants provided data at 3 months.

The intervention description characterizes PST as a standardized cognitive-behavioral treatment, delivered by trained therapists in 6 1-hour home sessions. ST is described as a similarly structured psychological treatment intended to control for nonspecific treatment effects. The remaining ST description is cut off, and further treatment details are not reported in the registry record.

Participants were described as having bilateral AMD and visual acuity worse than 20/70 in the better eye. The brief summary places recruitment at retina clinics at Wills Eye Institute in Philadelphia. The record identifies the study as single-masked but does not say which participants or study personnel were masked. The randomization method, allocation concealment procedures, and the trial's original sample-size 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.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: 106 PST and 112 ST participants provided data at 3 months.

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 Targeted Vision Function (TVF), measured over 3-Months.

The primary endpoint was Targeted Vision Function (TVF), assessed at 3 months and expressed in units on a scale. It was designed to capture difficulty with vision-related activities that individual participants considered important. At baseline, participants completed the Activities Inventory, which asks about the value and difficulty of vision-function goals and the tasks needed to accomplish them. For TVF, participants selected up to 4 valued goals and rated how difficult each was. Their average difficulty rating formed the score; they rated the same goals again at outcome assessment.

The score ranges from ratings of not difficult to impossible for goals considered important. Higher average scores indicate greater disability. This individualized approach makes TVF relevant to daily tasks that matter to participants, rather than treating every vision-related activity as equally important. It also means that the endpoint reflects participants' ratings of selected goals, not a direct measure of visual acuity.

The registry reports 3-month TVF means for participants in both groups. The endpoint description explains its intended meaning, but further information about how missing goal ratings were handled or whether endpoint scoring changed during the trial is not reported in the registry record.

7. Results Posted in the Registry

• Supportive Therapy: mean 2.14 versus 2.18 on control; difference -0.04 (95% CI -2.59 to 2.51), standardized effect d = -0.00; z-test p = 0.975.

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.

Vision-related Quality of Life

• 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:
  • Supportive Therapy: mean 65.20 versus 66.60 on control; difference -1.40 (95% CI -5.53 to 2.73), standardized effect d = -0.09; z-test p = 0.506.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Supportive Therapy and the control. The interval gives the range of effects that remain compatible with the data.

The posted secondary outcome concerns vision-related quality of life at 3 months. The registry lists this outcome twice among secondary outcomes, while a detailed result is posted for one secondary endpoint. Across the primary and secondary results, the broad pattern is of uncertain between-group comparisons: the posted adjusted intervals include zero, and the summary-based primary comparison is also imprecise. This contextual similarity does not make the endpoints interchangeable, since TVF and vision-related quality of life measure different aspects of vision-related experience.

Secondary outcomes can help describe whether a primary result is accompanied by a broader pattern across related measures. They do not independently settle the primary question, particularly when there are several outcomes and the record does not describe how multiplicity was handled. When many outcomes are examined, the chance of obtaining an apparently notable result by chance can increase. Adjusting interpretation or inference for multiple comparisons can help manage that concern, but a multiplicity plan is not reported in the registry record.

Secondary findings are generally less central to the trial's principal question than the designated primary endpoint. Their interpretation can also be affected by missing outcome data, analytic choices, and whether outcome definitions and analysis plans were set in advance. Those details are not reported in the registry record. The secondary result can therefore be treated as supporting context, not as a separate confirmatory conclusion. No conclusion about consistency across every intended secondary measure is possible from the limited posted results.

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

• Vision-related Quality of Life
• Vision-related Quality of Life

9. Statistical Interpretation

10. Statistical Methods Explained

Two forms of comparison appear in the posted primary results: a simple recalculation from group summary statistics and a sponsor-reported ANCOVA. They answer related, but not identical, statistical questions. Understanding that distinction is important when interpreting the estimates and their uncertainty.

What does a mean difference represent?

A mean difference compares the average outcome in one group with the average in another. For a continuous endpoint such as TVF, the group means describe the central outcome level among participants with recorded measurements. The sign depends on the order in which the groups are subtracted. In this analysis, the simple recalculation reports a difference of -0.04 for Supportive Therapy relative to Problem-Solving Treatment, based on means of 2.18 and 2.14, respectively.

A mean difference is expressed on the outcome's original scale. It does not, by itself, communicate how variable individual outcomes are or how precisely the group contrast is estimated. The spread of observations and the number of participants with outcome data contribute to that precision. A small estimated difference can be compatible with considerable uncertainty, particularly when variability is substantial.

What is the simple recalculation?

The recalculated comparison uses the posted group summaries and is unadjusted. It does not account for baseline values or other covariates. Its estimate is -0.04, its 95% confidence interval is -2.592 to 2.512, and its p-value is 0.975. This is a straightforward comparison based on the summary data, not a reconstruction of the trial's full statistical analysis. It should not be mistaken for an adjusted model or for a result using participant-level data.

The registry reports standard errors for the primary group means. A standard error describes the uncertainty in an estimated mean, rather than the spread of individual participants' scores. The recalculation uses the posted summary information to estimate uncertainty around the group contrast. It cannot incorporate analytic choices or information that are not present in those summaries.

What does ANCOVA add?

ANCOVA is a regression-based method that compares groups while adjusting for one or more covariates. In a clinical trial, baseline outcome values are often relevant covariates, but the specific covariates used in this ANCOVA are not reported in the registry record. The registry reports a net mean difference of 0.079, with a 95% confidence interval from -0.14 to 0.29 and p = .47. Because the adjustment details are unavailable, the estimate should be described as the posted ANCOVA result rather than attributed to a particular adjustment set.

Adjustment can change both the estimated contrast and its precision. The adjusted and unadjusted estimates may differ because they use different statistical approaches and may target a contrast conditioned on covariates rather than a raw difference between group means. Their numerical differences do not, by themselves, establish an error or an inconsistency.

How should the confidence interval and p-value be read?

A confidence interval gives a range of values compatible with the data under the analysis assumptions. It conveys precision and the direction and scale of effects that remain plausible under that model. An interval spanning zero indicates that the analysis does not distinguish the estimated contrast from no difference at the stated confidence level. It does not demonstrate equivalence, and it does not show that every value in the interval is equally likely.

A p-value describes how unusual the observed data, or data more extreme, would be under a specified null model and its assumptions. It is not the probability that the null hypothesis is true, nor does it quantify the probability that a treatment is beneficial. The recalculated p-value is 0.975; the ANCOVA p-value is .47. Both are reported as results of their respective analyses and should be considered alongside their confidence intervals, not as stand-alone measures of importance.

Why does over-dispersion matter?

In general, over-dispersion means that outcomes vary more than a statistical model expects. When a method assumes less variability than the data actually contain, standard errors can be too small and confidence intervals too narrow, making results appear more precise than they are. The posted TVF results are treated as continuous means, but the registry record does not report an over-dispersion assessment or a specific model diagnostic. It is therefore not possible to say whether over-dispersion was assessed or affected the ANCOVA.

For continuous outcomes, variability among individual observations and uncertainty in group means are distinct quantities. The TVF arm summaries report standard errors, while the registry's adjustment details and participant-level distributions are not available. Statistical interpretation should therefore remain anchored to the analyses and uncertainty intervals actually reported, without assuming additional diagnostics or modeling choices.

11. Confidence Intervals and Precision

Precision describes how tightly an analysis estimates a group contrast. The primary unadjusted recalculation gives a mean difference of -0.04 and a 95% confidence interval from -2.592 to 2.512. This interval spans zero and a broad range of possible contrasts on the TVF scale. Its width indicates that the posted summary comparison does not pin down the magnitude or direction of the difference with much precision.

The sponsor-reported ANCOVA estimate is 0.079, with a 95% confidence interval from -0.14 to 0.29. This interval also spans zero, while being much narrower than the unadjusted recalculation interval. Because the analyses differ, their intervals should be interpreted separately. The record does not give enough detail about the ANCOVA covariates or model specification to explain fully why its precision differs.

A confidence interval is not a range that contains a fixed proportion of individual participant outcomes. It concerns uncertainty about an estimated group contrast under the assumptions of the analysis. Nor does an interval that includes zero prove that the groups are identical. It indicates that the data and method do not precisely separate the estimate from a null difference.

The secondary quality-of-life ANCOVA estimate also has a 95% interval that includes zero, from -1.96 to 2.77. That provides a similarly cautious picture for the reported adjusted comparison, but it does not narrow uncertainty about TVF itself. Missing outcome handling, the ANCOVA specification, and further information about outcome distributions are not reported in the registry record, limiting a fuller assessment of precision.

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 are participants per arm, not total event counts. They are illustrations based on the observed effect and are not the trial's original design calculations. 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 as the stated fraction of the observed effect.

The 80% power column represents the number of participants per arm that the illustration estimates would be needed to have an 80% chance of detecting the specified effect under those assumptions, if the effect and model assumptions were correct. The 90% power column describes the corresponding planning target for a 90% chance. These are planning probabilities under an assumed effect, not guarantees about what a future study would find. The trial's own sample-size justification and original design assumptions are not reported in the registry record.

The entries for 100% of the observed primary effect are 911,906 participants per arm for 80% power and 1,220,783 participants per arm for 90% power. Planning for 75% of the observed effect gives 1,621,166 per arm for 80% power and 2,170,281 per arm for 90% power. Planning for 50% of the observed effect gives 3,647,623 per arm for 80% power and 4,883,132 per arm for 90% power. Smaller effects require more participants to detect under the same assumptions, which is why the illustrated requirements rise as the effect fraction gets smaller.

The curve presents the same planning relationship across assumed effects. These calculations are not observed power: they do not describe the probability that the completed trial detected an effect after its results were known. They are scenario-based illustrations using the observed estimate, which is itself uncertain. The unusually large estimates should be read in that context, rather than as a statement about the actual enrollment or a recommended future sample size.

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, Δ is the difference in means to detect and σ is the common standard deviation.

Worked calculation for this trial

1. Observed difference Δ = -0.040; pooled standard deviation σ = 9.641.
2. n = 2 × (1.96 + 0.842)² × 9.641² / -0.040² = 911,905.5, rounded up to 911,906 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
delta <- -0.04                      # observed difference in means
sigma <- 9.64086671991222                     # pooled standard deviation
n_per_arm <- function(power, frac = 1) {
  ceiling(round(2 * (z_a + qnorm(power))^2 * sigma^2 / (delta * frac)^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 (t-based, so very slightly larger)
cat("power.t.test:", round(power.t.test(delta = delta, sd = sigma, power = 0.80)$n, 1), "\n")

# Simulation: exact two-sample t-test using simulated means and variances
set.seed(572039)
n    <- n_per_arm(0.80)
reps <- 20000
m_diff <- rnorm(reps, delta, sigma * sqrt(2 / n))             # difference in sample means
s2     <- sigma^2 * rchisq(reps, 2 * (n - 1)) / (2 * (n - 1))   # pooled sample variance
tstat  <- m_diff / sqrt(s2 * 2 / n)
cat(sprintf("Simulated power at n = %s per arm: %.3f (%d replicates)\n", format(n, big.mark = ","), mean(abs(tstat) > qt(0.975, 2 * (n - 1))), reps))

Output from running the code

100% of observed effect: n per arm = 911,906 (80% power), 1,220,783 (90% power)
 75% of observed effect: n per arm = 1,621,166 (80% power), 2,170,281 (90% power)
 50% of observed effect: n per arm = 3,647,623 (80% power), 4,883,132 (90% power)
power.t.test: 911906.5 
Simulated power at n = 911,906 per arm: 0.799 (20000 replicates)

The simulation draws many sets of group means and variances at the planned sample size and counts how often the two-sample t-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 registry's sponsor-reported primary analysis is an ANCOVA with a net mean difference of 0.079, a 95% confidence interval from -0.14 to 0.29, and p = .47. The simple recalculation from posted arm summaries gives a mean difference of -0.04, a 95% confidence interval from -2.592 to 2.512, and p = 0.975. Both intervals include zero, but the estimates and precision differ.

The distinction is methodological. The recalculation is an unadjusted comparison based on group summaries. ANCOVA is an adjusted model, and the registry does not report its covariates or full specification. Different adjustment and estimation procedures can produce different estimates and standard errors. The adjusted estimate should therefore be read as the sponsor's reported model result, not as a direct replication of the unadjusted recalculation.

The record does not identify a discrepancy between the reported analysis and the summary-based recalculation. However, the information available does not permit a detailed audit of the ANCOVA or a reconstruction of it from participant-level data. The two results are best presented with their methods distinguished, rather than choosing one as a replacement for the other.

15. Multiplicity, Missing Data and Analysis Populations

Multiplicity arises when a study evaluates several outcomes, time points, subgroups, or statistical comparisons. Each additional test can increase the chance that at least one result appears unusual under a null model, even when no underlying effect exists. Common approaches include prespecifying a hierarchy of outcomes or adjusting thresholds or intervals to account for multiple testing. The registry record does not report a multiplicity strategy, so it is not possible to determine whether the posted secondary analyses were intended as confirmatory or exploratory, or how any multiplicity was addressed.

The registry names Vision-related Quality of Life twice among the secondary outcomes and reports a detailed result for that endpoint. It also reports a total of 4 outcome measures and 2 analyses. The relationship between those counts and the full set of planned, collected, and analyzed outcomes is not explained. Prespecified outcome definitions, a statistical analysis plan, and details about selection of reported analyses are not reported in the registry record.

Missing data are relevant because the primary endpoint population is described as 106 PST and 112 ST participants with 3-month data. The enrollment field records 241, while the brief trial description says 120 participants were randomized to each arm. The registry record does not provide a complete participant flow or explain the difference between enrollment, stated arm numbers, and the observed primary endpoint counts. The extent and reasons for missing outcome data are not reported in the registry record.

The record also does not say whether the primary analysis used all randomized participants, only participants with observed outcomes, or another analysis population. It does not explain whether missing values were imputed, whether sensitivity analyses were conducted, or how participants who did not complete treatment were handled. These omissions limit assessment of the impact of missingness and the robustness of the reported estimates.

Although the reported method is ANCOVA, the registry does not specify the covariates, model assumptions, or handling of incomplete covariate and outcome data. Those details matter for reproducing and evaluating the adjusted estimate. Without them, the posted result can be described accurately, but the analysis cannot be fully reconstructed from the registry record.

16. Safety Results

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

Other (non-serious) adverse events

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

Most frequent serious adverse event terms

Serious events with the highest arm-level rate: fall (6.7%), pneumonia (4.2%), depression (4.1%), cardiac event (3.3%), congestive heart failure (2.5%).

The registry reports serious adverse events affecting 35 of 121 participants at risk in the Problem-Solving Treatment group and 29 of 120 at risk in the Supportive Therapy group. It reports zero other adverse events affecting participants in either group. These counts describe what was recorded in the registry; they do not by themselves establish that treatment caused an event or that the groups differed in safety.

The listed serious adverse event terms include Fall (6.7%), Pneumonia (4.2%), Depression (4.1%), Cardiac event (3.3%), Congestive Heart Failure (2.5%), and Dehydration/anemia (2.5%). The record does not give further detail in the registry record about event timing, severity beyond the serious classification, or the basis for attributing events to either treatment. Death counts are not reported in the registry record.

Serious adverse event counts should be read alongside the number at risk and the study context, not as a complete account of all possible harms. The registry information alone does not support a causal comparison or a conclusion about safety differences.

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

The posted results describe a randomized comparison of Problem-Solving Treatment and Supportive Therapy in people with age-related macular degeneration. The primary endpoint was a participant-rated measure of difficulty with personally selected, valued vision-function goals at 3 months. The recorded group means were close, and both the unadjusted recalculation and the sponsor-reported adjusted analysis have confidence intervals that include zero.

These findings do not establish that the two treatments are equivalent. Equivalence requires a suitable design and a prespecified margin defining how small a difference must be to count as practically equivalent. Such a margin is not reported in the registry record. Nor do the results prove that there is no difference. The unadjusted interval is wide, leaving considerable uncertainty about the size and direction of a possible contrast.

The results also do not establish that either treatment improves vision function compared with no treatment or usual care. The trial compared two active psychological treatments, and there is no untreated or usual-care comparison in the posted design description. The primary outcome is a measure of self-rated functional difficulty, not visual acuity or restoration of vision.

Interpretation is limited to the recorded follow-up and analyzed participants. The primary outcome population includes 106 PST and 112 ST participants with data at 3 months. Missing data handling and the analysis population for the adjusted result are not reported in the registry record. Further details about the ANCOVA and the trial's original sample-size plan are also not reported.

The appropriate statistical reading is therefore restrained: the registry records no clearly separated group result on the primary endpoint at 3 months, while uncertainty remains. This is a description of the posted analyses, not medical advice or a judgment about which treatment an individual should receive.

18. Why This Trial Matters Statistically

This trial illustrates why an endpoint's meaning matters as much as its numerical result. TVF focuses on participant-selected activities that are important but difficult, making it distinct from a visual-acuity measure. A statistically analyzed outcome is most interpretable when readers can understand what it measures, how it was scored, and which population contributed data.

It also shows the value of separating an adjusted model result from an unadjusted summary comparison. The ANCOVA estimate and the simple recalculation differ in estimate and precision, but they arise from different methods. Naming that distinction prevents a reader from treating the two analyses as duplicate calculations or assuming that one can be reproduced from the other without the model details.

The confidence intervals demonstrate why a non-small p-value is not the same as evidence of no difference. The primary unadjusted interval is broad, while the sponsor-reported adjusted interval is narrower; both include zero. Considering the estimates, intervals, and analysis methods together gives a more informative account than focusing on a p-value alone.

The record also makes visible questions that matter in clinical-trial interpretation: who contributed outcome data, how missingness was handled, how secondary outcomes were selected and interpreted, and what assumptions guided planning. These details are not fully reported in the registry record. Their absence does not invalidate the results, but it limits independent evaluation and replication.

19. 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.

Several trial-specific limitations follow from what is and is not reported. The primary outcome results are based on 106 PST and 112 ST participants with 3-month data, while the registry records enrollment of 241 and the brief description states 120 participants per arm. The record does not explain this difference or provide a full participant flow, reasons for missing outcomes, or the missing-data method.

The trial is described as single-masked, but the identity of those masked is not reported. The randomization procedure and allocation concealment are also not reported. These omissions prevent a full assessment of protections against selection and assessment biases.

The sponsor-reported analysis is ANCOVA, but the registry does not list the covariates or detailed model specification. The simple recalculation does not adjust for baseline values and cannot reproduce an analysis based on participant-level data. The original sample-size justification and design assumptions are not reported; the planning illustrations based on the observed effect should not be treated as the trial's design rationale.

The record identifies secondary outcomes but does not report a multiplicity approach, and the relationship between the listed outcome count and the posted secondary results is unclear. Further details about the intervention and the comparator are limited, including a truncated description of ST. These gaps constrain interpretation of the treatment contrast, secondary findings, and safety information. They do not establish that a particular bias occurred; they mark areas where the registry record does not permit a fuller evaluation.

20. Related Tutorials

21. Related Statistical Calculators

22. Sources

• ClinicalTrials.gov: Improving Function in Age-Related Macular Degeneration — NCT00572039.
• Registry record last updated 2025-04-15; status last verified 2025-04.
• Results first posted 2015-10-15.
• 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.

23. Record Summary

NCT00572039 is a completed randomized, single-masked, parallel-group phase 3 trial sponsored by Thomas Jefferson University. It compared Problem-Solving Treatment with Supportive Therapy for people with age-related macular degeneration. The registry records enrollment of 241, with a study start of 2005-08 and primary completion in 2013-06.

The primary endpoint was Targeted Vision Function at 3 months, a measure of difficulty with personally selected, valued vision-related goals. Data were reported for 106 PST and 112 ST participants. The simple unadjusted recalculation and the sponsor-reported ANCOVA both have confidence intervals that include zero, but they use different methods and have different precision.

Missing-data handling, the analysis population, ANCOVA details, multiplicity procedures, and the original sample-size justification are not reported in the registry record. Serious adverse events were recorded in both groups; the registry information does not establish causation or a comparative safety conclusion.