Oligonucleotide Ligation Assay (OLA) Resistance Study: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Oligonucleotide Ligation Assay (OLA) Resistance Study (NCT01898754), a randomized na study. The primary outcome is Number of Participants With Virologic Failure With OLA-guided ART vs. Standard of Care.

2. Clinical Question

3. Trial Design

This was a randomized, parallel-group trial with no masking reported. The registry lists phase as NA, total enrollment as 991, and status as completed. The recorded start is 2013-05 and primary completion is 2016-12. The sponsor was the University of Washington. The trial was conducted in the context of HIV care at the PEPFAR Coptic Hope Center in Kenya, as described in the registry summary.

One group was assigned pre-ART OLA testing. The test was intended to identify resistance at five pol codons: K103N, V106M, Y181C, G190A, and M184V. The other group received standard of care according to Kenya guidelines and was to be offered OLA resistance testing after 12 months. The intervention is classified in the record as “OTHER”; further operational details about how clinicians used test results, treatment algorithms, adherence support, or implementation fidelity are not reported in the registry record.

The primary endpoint was a comparison of the rates of viral non-suppression >400 copies/mL between groups at 12 months. The record also lists secondary questions involving virologic failure in participants with transmitted drug resistance, participants with prior or ongoing ART use, resistance prevalence, medical resource utilization, longer-term cost-effectiveness, and low-level resistance detected by PYRO but not by OLA. These are the registered outcome topics; the registry record does not provide results for each of them in the posted outcome data.

Although the design is identified as randomized, details such as the randomization sequence generation, allocation concealment, stratification, and implementation procedures are not reported in the registry record. The original sample-size justification and other protocol-level statistical assumptions are also not reported in the registry record. Those omissions limit what can be inferred about design choices, but they do not alter the interpretation of the posted event counts and comparison estimate.

4. Participant Flow and Accounting

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

Overall Study

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Number of Participants With Virologic Failure With OLA-guided ART vs. Standard of Care, measured over 12 months.

The primary endpoint counts participants with virologic failure, described as viral non-suppression with viral load >400 copies/mL at 12 months. It is a binary outcome: for the endpoint, each participant is classified according to whether the defined failure event occurred. The registry title describes the result as a number of participants, while the endpoint description frames the intended comparison as rates between the randomized groups.

A binary virologic outcome gives a direct group-level summary of how often the endpoint event was recorded during the specified assessment period. It is clinically interpretable as a count and can also be compared between groups using measures such as a risk difference, risk ratio, or odds ratio. These measures answer related but not identical questions: the risk difference expresses the contrast in event proportions, whereas the ratio measures compare event frequency on a relative scale. General statistical explanation: reporting both counts and a comparative estimate helps readers distinguish the observed number of events from the estimated size and direction of the group contrast.

The posted primary outcome lists 34 events among 400 participants in the OLA group and 39 among 403 participants in standard care. The record’s stated endpoint time is 12 months. Details about missing viral-load measurements, the exact handling of participants without an endpoint assessment, and any rules for classifying borderline or repeated measurements are not reported in the registry record.

7. Results Posted in the Registry

• Pre-ART Oligonucleotide Assay (OLA): 8.5% (34/400) versus 9.7% (39/403) on control; risk difference -1.2 percentage points (95% CI -5.2 to 2.8), risk ratio 0.88 (0.57–1.36); unadjusted two-proportion z-test p = 0.562.

8. Statistical Interpretation

9. Statistical Methods Explained

What does the risk difference describe?

General statistical explanation: a risk difference compares the proportion experiencing an event in one group with the proportion in another. A value below zero for the OLA-minus-standard-care contrast indicates a lower observed event proportion in the OLA group; a value above zero would indicate a higher proportion. The posted estimate is negative, but its confidence interval crosses zero. That means the data do not rule out either direction of difference at the interval’s confidence level.

The registry also gives proportions for the two groups, alongside the participant counts. Counts provide an intuitive account of how many endpoint events were recorded, while proportions account for the different denominators. The risk difference expresses their contrast on an absolute scale. It is often useful for understanding the size of a group difference in the units of event probability, but it remains an estimate subject to sampling variability.

How do ratios differ from the risk difference?

General statistical explanation: a risk ratio compares the event proportion in one group with the event proportion in another. An odds ratio compares the odds of the event rather than the proportions themselves. Both are relative measures, whereas the risk difference is absolute. Relative measures can appear larger or smaller depending on the baseline frequency of the event, so their interpretation should be tied to the observed event counts and the clinical question.

The registry posts risk-ratio and odds-ratio estimates and confidence intervals for the primary comparison. Their intervals include the ratio value corresponding to no group difference. Thus, these relative summaries also leave substantial uncertainty about the direction and size of any difference. The record does not report an adjusted model, so the posted measures should be read as unadjusted comparisons rather than as estimates adjusted for listed prognostic factors.

What does the p-value say?

General statistical explanation: a p-value evaluates how compatible the observed data are with a specified null hypothesis under the assumptions of the statistical test. It is not the probability that the treatment effect is zero, and it does not measure the probability that a result will replicate. The posted unadjusted p-value is 0.562. That value does not provide evidence of a clear group separation in this comparison, but it cannot establish equivalence or prove that no difference exists.

Statistical significance and practical importance are also distinct. A small p-value, if present, would not by itself establish that an effect is large or important. Conversely, a non-small p-value can occur when the estimate is imprecise. Here, the interval spanning negative and positive differences shows why the point estimate and p-value should be read alongside uncertainty rather than in isolation.

Why does the endpoint’s binary form matter?

General statistical explanation: a binary endpoint reduces each participant’s outcome to whether the specified event occurred during the endpoint window. This makes event counts and group proportions straightforward to communicate, but it does not preserve all information about viral-load values, timing of failure, duration of suppression, or repeated measurements. The primary endpoint is stated at 12 months, so the posted comparison is a snapshot at that time rather than a time-to-event analysis.

The trial’s other stated outcomes address additional questions, including virologic failure in selected subgroups, prevalence of transmitted drug resistance, and resource use. Those outcomes could add context, but their methods and results are not detailed in the primary comparison. The registry record lists 11 outcome measures overall and seven posted secondary outcomes; it does not report posted analysis results for each secondary question.

What can be said about the analysis population?

The primary-outcome counts use denominators of 400 and 403, while total enrollment is 991. The registry also lists different group counts at risk in the safety information. These figures should not be assumed to refer to identical analysis populations or follow-up sets. The participant flow, the numbers randomized to each group, the handling of missing endpoint measurements, and whether an intention-to-treat or per-protocol population was used are not reported in the registry record.

General statistical explanation: missing outcomes can affect estimates if missingness is related to a participant’s prognosis, group assignment, or observed outcome. A complete-case analysis, an imputation approach, and other strategies can make different assumptions. Because the registry record does not describe its missing-data methods, readers cannot determine which assumptions underlie the posted primary comparison.

The registry records zero posted analyses, while it does post a primary comparison and outcome counts. No additional inferential model specifications, multiplicity adjustments, or analysis-population definitions are reported. The available statistical interpretation should therefore remain focused on the posted unadjusted comparison and its stated uncertainty.

10. Confidence Intervals and Precision

The risk-difference confidence interval runs from a negative value to a positive value and includes zero. This indicates that the posted primary estimate is not precise enough to identify the direction of the group difference with confidence. The interval represents a range of values compatible with the data under the method’s assumptions; it is not a range containing a fixed proportion of individual participants’ outcomes, nor is it a guarantee that the true effect lies within it.

For the ratio measures, the registry’s confidence intervals include the no-difference value for a ratio. The risk-difference and ratio intervals therefore tell a consistent broad story: the point estimates favor a lower observed event frequency in the OLA group, but the data also remain compatible with no difference and with a difference in the opposite direction. The unadjusted p-value of 0.562 is consistent with that uncertainty.

Precision depends on the amount of information in the data, including the number of participants with outcome information and the frequency of events. The posted primary groups contain 400 and 403 participants, with 34 and 39 events, respectively. These counts describe the endpoint data used in the displayed comparison; the record does not explain why the denominators differ from total enrollment or provide a detailed account of missing outcome assessments.

A confidence interval is not a measure of clinical importance by itself. Interpretation also depends on what size of difference would matter for the research question, and that threshold is not reported in the registry record. The uncertainty should therefore be described directly rather than converted into a claim that the groups are equivalent or that a meaningful effect has been excluded.

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 participants and not total events. They are illustrative calculations based on the observed effect, rather than the original trial’s own design justification. The planning assumptions are a two-sided significance level of 0.05, equal allocation between two arms, and a normal approximation. The assumed effect to detect is set at the stated fraction of the observed effect. The trial’s original sample-size assumptions and justification are not reported in the registry record.

The 80% power column describes the number of participants per arm that the planning calculation associates with an 80% chance of detecting the specified effect under those assumptions, if that effect were the true one. The 90% power column describes the corresponding per-arm number for a 90% chance. These are design-planning probabilities for a hypothetical future study under a model; they are not probabilities that the posted trial’s result is correct.

The illustration gives 9,355 participants per arm for 80% power and 12,524 per arm for 90% power when planning for 100% of the observed effect. For 75% of the observed effect, the corresponding figures are 16,874 and 22,589 participants per arm. For 50% of the observed effect, they are 38,511 and 51,554 participants per arm. Smaller effects require more participants to detect under the stated assumptions, because they are harder to distinguish from random variation.

These calculations use the observed effect as their starting point, so they should be read as planning illustrations rather than as evidence about what the completed trial was powered to detect. This is not observed power.

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 = 0.097; treatment proportion p₁ = 0.085; average p̄ = 0.091.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.4065 = 0.7968.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.4064 = 0.3421.
4. n = (0.7968 + 0.3421)² / (-0.0118)² = 9,354.9, rounded up to 9,355 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 <- 0.0967741935483871                          # control proportion
p1 <- 0.085                          # 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(1898754)
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 = 9,355 (80% power), 12,524 (90% power)
 75% of observed effect: n per arm = 16,874 (80% power), 22,589 (90% power)
 50% of observed effect: n per arm = 38,511 (80% power), 51,554 (90% power)
power.prop.test: 9354.9 
Simulated power at n = 9,355 per arm: 0.802 (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 11 outcome measures and seven posted secondary outcomes. The secondary topics include virologic failure in participants with transmitted drug resistance or prior or ongoing ART use, resistance prevalence by sequencing and OLA, failure within a subgroup with transmitted resistance, medical resource utilization, longer-term cost-effectiveness, and low-level resistance detected by PYRO but not OLA. The registry record does not provide results for each listed secondary topic in the posted outcome information.

When a study examines multiple endpoints, subgroups, or contrasts, the chance of at least one apparently unusual result can increase. General statistical explanation: multiplicity procedures can control that risk, but they may also change the threshold or uncertainty used for individual findings. The registry record does not report whether multiplicity was addressed, whether a hierarchy of outcomes was prespecified, or whether the secondary questions were exploratory. No multiplicity adjustment should therefore be assumed.

The registry contains no posted analysis records, and it does not describe the statistical analysis plan in sufficient detail to identify modeling choices, covariate adjustment, or subgroup procedures. The primary comparison is identified as unadjusted in the posted comparison. The record does not report whether alternative models or sensitivity analyses were conducted.

Missing-data methods and analysis populations are also not reported in the registry record. The primary endpoint denominators are 400 and 403, whereas total enrollment is 991. The record does not explain the reasons for this difference, give a participant disposition by group, or state whether all randomized participants were included in the primary analysis. It also does not identify how missing viral-load values were handled or define a per-protocol population.

These omissions do not change the observed counts, but they limit assessment of whether the posted comparison is robust to missing outcomes, analytic choices, or multiple testing. Conclusions about secondary outcomes, subgroup effects, and the analysis population should be limited to what is explicitly recorded.

14. 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

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: pneumonia (1.0%), tuberculosis (1.0%), anemia (0.8%), stevens johnson syndrome (0.8%), blocked fallopian tubes (0.6%).

The registry lists serious adverse events for 492 participants at risk in the OLA group and 495 in standard care. It records 39 participants with a serious adverse event in the OLA group and 34 in standard care. For other adverse events, the corresponding counts at risk are 492 and 495, with 32 and 26 participants affected. The death counts are 34 and 43, respectively. These are the registry’s posted counts; the record does not provide a comparative statistical analysis of these safety outcomes.

The serious-event terms listed as most frequent include pneumonia and tuberculosis, each reported as 1.0%; anemia and Stevens Johnson Syndrome, each 0.8%; and blocked fallopian tubes and peptic ulcer disease, each 0.6%. The registry record does not describe attribution to the intervention, event severity criteria beyond the serious-event category, or the timing and duration of events.

These figures should be read as descriptive safety reporting, not as proof that the intervention caused or prevented an event. The safety denominators differ from the primary-outcome denominators, and the record does not explain the analysis populations or participant flow. Without further definitions and comparative analyses, counts alone cannot establish a causal safety difference between groups.

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

The posted results show 34 primary endpoint events among 400 participants in the pre-ART OLA group and 39 among 403 in standard care. The estimated risk difference is negative, but the confidence interval includes zero and spans differences in both directions. The unadjusted p-value is 0.562. Together, these statistics indicate that the registry’s primary comparison is uncertain and does not show a clear statistical separation between the groups.

This result does not establish that the groups are equivalent. An equivalence conclusion requires a prespecified margin defining differences small enough to be considered practically equivalent, along with an analysis designed for that question. Such a margin and an equivalence analysis are not reported in the registry record. Nor does the result establish that OLA-guided care has no effect: the interval is compatible with more than one direction and magnitude of effect.

The finding also does not answer every aim described in the registry summary. The primary comparison is a binary outcome at 12 months. It does not by itself describe the duration of suppression, changes over time, long-term cost-effectiveness, or the listed subgroup questions. Although these subjects appear among the registered outcomes, their methods and results are not reported in the posted primary comparison.

Interpretation is further constrained by the absence of detailed information on missing outcomes, analysis populations, adjustment, and participant flow. The primary endpoint denominators are smaller than total enrollment, and the registry does not explain how the analyzed participants were selected. The observed outcome counts remain informative as descriptive results, but the record does not support conclusions about why participants were absent from those denominators.

The most defensible reading is therefore narrow: the posted data estimate a small negative risk difference for 12-month virologic failure, with uncertainty that includes no difference and the opposite direction. That is a description of the reported comparison, not medical advice or a claim that either strategy works or does not work.

16. Why This Trial Matters Statistically

This trial raises a statistical question that is common in diagnostic-strategy research: whether adding resistance information before treatment decisions changes a patient-level clinical outcome compared with standard care. The endpoint is not the test result itself, but virologic failure at 12 months. That distinction matters because an assay can identify resistance while the relevant trial question concerns whether using that information changes the measured outcome.

The comparison also involves timing. The standard-of-care group was to be offered OLA resistance testing after 12 months, so the contrast is early test-guided care versus the stated standard-care approach during the primary endpoint period. It should not be described as testing versus permanent absence of testing. The intervention’s interpretation depends on the use of test results in clinical decisions, details of which are not reported in the registry record.

The posted result illustrates why an estimate must be read together with its precision. The event counts and negative risk-difference estimate point in one direction, but the confidence interval includes zero and values on either side. The unadjusted p-value does not establish equivalence, and the point estimate alone cannot settle the question. This is particularly important when translating an efficacy-style comparison into decisions about implementing a testing strategy.

The registry also names resource utilization and long-term cost-effectiveness among the research questions. Those outcomes involve different data and methods from the primary binary endpoint. Statistical conclusions about virologic failure cannot substitute for economic results, and the registry record does not report those results in the posted comparison. The value of this record for statistical interpretation is its explicit endpoint, group counts, and uncertainty, alongside clear limits on what can be concluded.

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.

Several trial-specific limitations in the registry record constrain interpretation. Total enrollment is 991, while the posted primary endpoint denominators are 400 and 403. The safety counts at risk are 492 and 495. The record does not explain these denominator differences, provide participant flow by randomized group, or state the analysis population used for each outcome. Consequently, it is not possible to determine from the registry record whether the primary comparison includes all randomized participants or how missing endpoint data were handled.

The record identifies the study as randomized and parallel, with no masking. It does not report how the random sequence was generated, whether allocation was concealed, whether randomization was stratified, or how treatment assignments were implemented. These details are relevant to evaluating the design, but their absence from the registry record does not establish that the procedures were inadequate.

The primary comparison is posted as unadjusted. No adjusted estimate, model specification, sensitivity analysis, or multiplicity procedure is reported. The registry lists multiple secondary questions, but results and analysis details for those questions are not fully available in the posted outcome information. This limits assessment of subgroup findings and of whether the primary comparison was interpreted within a prespecified testing hierarchy.

The registry summary describes goals related to durable viral suppression and cost-effectiveness, but the posted primary endpoint is a 12-month binary failure measure. The record does not provide detailed results on repeated viral-load measurements, longer-term outcomes, or the economic questions. It also does not report the original sample-size justification or protocol-level statistical assumptions. These limitations mean that conclusions should be confined to the posted endpoint counts and comparison rather than extended to unreported outcomes or design details.

18. Related Tutorials

19. Related Statistical Calculators

20. Sources

• ClinicalTrials.gov: Oligonucleotide Ligation Assay (OLA) Resistance Study — NCT01898754.
• Registry record last updated 2025-03-12; status last verified 2025-03.
• Results first posted 2021-11-17.
• 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

The registry record describes a completed, randomized, parallel-group study of pre-ART OLA-guided care versus standard of care in Kenya, with no masking and phase listed as NA. Total enrollment is 991. The primary outcome is virologic failure at 12 months, defined as viral non-suppression >400 copies/mL.

The posted primary counts are 34 of 400 in the OLA group and 39 of 403 in standard care. The risk-difference estimate is negative, but its confidence interval includes zero; the unadjusted p-value is 0.562. Thus, the posted comparison is uncertain and does not establish a clear group difference or equivalence.

The record also lists secondary outcome topics and descriptive adverse-event counts, but it does not report detailed participant flow, missing-data methods, analysis populations, the original sample-size justification, or results for each secondary question. Interpretation should remain limited to the outcomes and statistical information explicitly recorded.