Prescription Opioid Addiction Treatment Study (POATS): Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Prescription Opioid Addiction Treatment Study (POATS) (NCT00316277), a randomized phase 3 study. The primary outcome is The Number of Participants Attaining Successful Opioid Use Outcome by Counseling Condition at End of Phase 1.

2. Clinical Question

3. Trial Design

The Prescription Opioid Addiction Treatment Study (POATS), registered as NCT00316277, was a Phase 3 randomized, open-label, parallel-group clinical trial sponsored by Mclean Hospital. The registry record lists the study as completed. Its dates are recorded as 2006-05 for study start and 2009-07 for primary completion. Enrollment was 653.

The study compared buprenorphine/naloxone with EMM against buprenorphine/naloxone with SMM. The trial’s stated purpose was to determine whether adding individual drug counseling to the prescription of buprenorphine/naloxone with standard medical management could improve treatment outcome. Because masking is listed as none, the study was open-label; the registry record does not report further details about concealment or how outcome assessment was conducted.

The registered study had multiple stages. The brief summary describes an initial four-week treatment with taper, followed by a 12-week stabilization treatment for those who did not respond successfully to the initial treatment, and long-term follow-up assessments at 1.5 years, 2.5 years, and 3.5 years after treatment. The primary endpoint is specified at 12 weeks, at the end of Phase 1. The registry’s Phase 2 follow-up outcome is reported at 24 weeks in the Phase 2 period, corresponding to 36 weeks into the study.

The primary outcome population is described as the 653 participants randomized to Phase 1, included in analysis. The two primary arm records list 329 participants for EMM and 324 for SMM. The registry record does not report the original design assumptions, allocation ratio rationale, a protocol-level sample-size justification, or detailed procedures for assignment. Randomization and parallel-group allocation are reported, but further implementation details are not.

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.

Phase 1 (12 Weeks)

Reasons for not completing this period:

Phase 2 (24 Weeks)

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: 653 study participants randomized to Phase 1 were included in analysis.

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 The Number of Participants Attaining Successful Opioid Use Outcome by Counseling Condition at End of Phase 1, measured over 12 weeks.

The primary endpoint counts participants who attained the registry’s definition of successful opioid-use outcome by the end of Phase 1. The timeframe is 12 weeks. Success required completing week 12 with self-reported opioid use on no more than 4 days in a month, no 2 consecutive opioid-positive urine test results, no additional substance use disorder treatment other than self-help, and no more than 1 missing urine sample during the 12 weeks.

This is a composite, rule-based outcome: a participant had to meet each listed condition to count as successful. It therefore reflects more than self-reported opioid use alone. It incorporates urine test results, treatment completion, additional treatment, and availability of urine samples. A count of participants meeting the full definition is straightforward to interpret as the number classified as successful under the registry’s criteria, but it does not show which individual criterion was or was not met by those who did not qualify.

The endpoint is used to summarize a clinically relevant treatment outcome over the specified period and to compare the counseling conditions. Its interpretation depends on the exact definition and the completeness and consistency of its component measurements. The registry record gives the endpoint definition but does not report further details about measurement procedures, adjudication, or how the individual components were combined in the analysis beyond the stated success rule.

7. Results Posted in the Registry

• Buprenorphine/Nx With SMM: 7.4% (24/324) versus 5.8% (19/329) on control; risk difference 1.6 percentage points (95% CI -2.2 to 5.4), risk ratio 1.28 (0.72–2.30); unadjusted two-proportion z-test p = 0.400.

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.

The Number of Participants Attaining Successful Opioid Use Outcome by Counseling Condition Phase 2, 8-week Posttreatment Follow-up

• 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:
  • Buprenorphine/Nx With SMM: 7.2% (13/180) versus 10.0% (18/180) on control; risk difference -2.8 percentage points (95% CI -8.6 to 3.0), risk ratio 0.72 (0.36–1.43); unadjusted two-proportion z-test p = 0.348.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Buprenorphine/Nx With SMM and the control. The interval gives the range of effects that remain compatible with the data.

The Number of Participants Attaining Successful Opioid Use Outcomes in Phase 1 by Chronic Pain Condition

• 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:
  • Success in Phase 1 Among Participants Without Chronic Pain: 4.7% (13/274) versus 7.9% (30/379) on control; risk difference -3.2 percentage points (95% CI -6.9 to 0.5), risk ratio 0.60 (0.32–1.13); unadjusted two-proportion z-test p = 0.107.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Success in Phase 1 Among Participants Without Chronic Pain and the control. The interval gives the range of effects that remain compatible with the data.

The Number of Participants Attaining Successful Opioid Use Outcomes in Phase 2 by Chronic Pain Condition

• 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:
  • Success in Phase 2 Among Participants Without Chronic Pain: 46.4% (98/211) versus 53.0% (79/149) on control; risk difference -6.6 percentage points (95% CI -17.0 to 3.9), risk ratio 0.88 (0.71–1.08); unadjusted two-proportion z-test p = 0.219.
• Interpretation:
  • The 95% confidence interval includes no difference, so these data do not show a statistically significant difference between Success in Phase 2 Among Participants Without Chronic Pain and the control. The interval gives the range of effects that remain compatible with the data.

The secondary results extend the primary question in two ways: they report a later counseling-condition comparison and explore outcomes by chronic pain condition. The posted pattern is not uniform in direction across these analyses. The Phase 2 counseling comparison has more successful outcomes in the EMM arm, whereas the primary Phase 1 comparison has more in SMM. The pain-condition comparisons also show different count patterns across phases, but their group sizes differ, so raw counts alone do not establish which condition had a higher outcome rate.

None of the available secondary confidence intervals excludes zero, and the listed unadjusted p-values do not provide clear evidence of a difference. These findings do not overturn the primary result: they address other timeframes or participant groupings and remain uncertain themselves. The registry record lists further secondary outcome topics, including outcomes by lifetime heroin use, but numerical results for those topics are not reported in the registry record.

Secondary analyses generally carry less weight than the primary analysis when they were not designated as the central trial question or when many outcomes and subgroups are examined. Testing multiple contrasts increases the chance that some estimates will look notable by chance. No multiplicity adjustment is reported, and the registry record does not provide a prespecified hierarchy or analysis plan for these outcomes. For that reason, secondary patterns are best treated as descriptive or hypothesis-generating rather than as independent confirmation of a treatment effect.

Comparisons by chronic pain status also differ from randomized treatment comparisons: pain status was not the assigned intervention. Any differences may reflect other group characteristics, and no adjusted or interaction analysis is reported. The secondary outcomes can help identify questions for further study, but they cannot establish that chronic pain or counseling condition caused the observed patterns.

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

• The Number of Participants Attaining Successful Opioid Use Outcome by Counseling Condition Phase 2, 8-week Posttreatment Follow-up
• The Number of Participants Attaining Successful Opioid Use Outcomes in Phase 1 by Chronic Pain Condition
• The Number of Participants Attaining Successful Opioid Use Outcomes in Phase 2 by Chronic Pain Condition
• The Number of Participants With and Without Any Lifetime Use of Heroin Attaining Successful Opioid Use Outcomes in Phase 1
• The Number of Participants With and Without Any Lifetime Use of Heroin Attaining Successful Opioid Use Outcomes in Phase 2

9. Statistical Interpretation

10. Statistical Methods Explained

The primary endpoint is a binary outcome: each participant is classified as attaining or not attaining the specified successful opioid-use outcome. The registry reports the number meeting the definition in each arm. For this kind of outcome, a comparison can be expressed as a difference between the proportions in the two groups, a relative risk, or an odds ratio. The posted comparison is described as a risk difference. The registry record gives no reported adjusted model or additional analysis specification, so the posted result should be understood as an unadjusted comparison.

What does an unadjusted comparison mean?

An unadjusted comparison contrasts the observed group outcomes without a reported statistical adjustment for other participant characteristics. Random assignment is intended to create comparable groups on average, but the registry record does not give the details of the randomization procedure or a covariate-adjusted analysis. An unadjusted estimate is therefore a description of the observed contrast in the analyzed groups, not a claim that every potential source of variation has been modeled.

The primary result includes 19 successful outcomes in EMM and 24 in SMM. The associated unadjusted p-value is 0.4. A p-value describes how unusual data at least as incompatible with a specified null hypothesis would be under the assumptions of the test; it is not the probability that the null hypothesis is true, nor does it measure the size or clinical importance of an effect. The registry record does not report a more detailed test specification, so interpretation should not assume one beyond the unadjusted comparison presented.

Why use a confidence interval?

A confidence interval communicates uncertainty around an estimated contrast. It gives a range of values compatible with the observed data under the method’s assumptions. For the primary risk difference, the interval spans zero. Zero represents no difference on the risk-difference scale, so the interval includes both no difference and values on either side. This indicates that the data do not pin down the direction or size of the contrast precisely. A confidence interval is not a guarantee that the true effect lies within its bounds, and it does not address limitations in the study design or outcome measurement.

How should subgroup comparisons be read?

The chronic-pain outcomes compare groups defined by pain condition, rather than the randomized counseling conditions. Such comparisons can be useful for describing outcome patterns, but randomization to EMM or SMM does not necessarily make participants with and without chronic pain comparable. The registry record does not report adjustment for baseline characteristics or a formal interaction analysis. Therefore, the chronic-pain comparisons should not be interpreted as evidence that pain status caused a difference in outcome or modified the counseling effect.

The Phase 2 counseling comparison concerns a later period and its own analysis population. It reports 180 participants in each counseling condition, with 18 and 13 successful outcomes, respectively. Its result is not interchangeable with the Phase 1 primary result: the endpoint timing and population differ. The registry does not report the details of how participants entered or were retained in the Phase 2 analysis beyond the counts and timeframe stated.

What does over-dispersion mean for these results?

Over-dispersion is a concern in count models when observed variability exceeds what the model assumes. It matters particularly for counts of repeated events or rates. The primary outcome is reported as participants attaining a binary outcome, rather than as a repeated-event rate, and the registry record gives no dispersion estimate or count-model analysis. There is therefore no basis to infer an over-dispersion adjustment from the posted record. In general, a model that ignores over-dispersion when it is present may understate uncertainty; whether that issue applies to a specific analysis depends on the outcome and model actually used.

The registry record reports 7 outcome measures and 0 analyses in its analysis listing. It gives outcome counts and unadjusted comparison summaries, but no additional analysis plans, covariate specifications, missing-data procedures, or multiplicity adjustments. The trial’s original statistical analysis plan is not reported in the registry record. Accordingly, the posted statistics support descriptive and unadjusted interpretation, while finer claims about model choice or analytic safeguards cannot be verified from the record.

11. Confidence Intervals and Precision

The primary comparison is uncertain. Its confidence interval for the risk difference extends across zero, which means the observed data are compatible with no difference and with effects in either direction. The interval is not concentrated on a single narrow set of values. The p-value of 0.4 is consistent with the lack of a clear separation from zero in the posted unadjusted comparison, but the interval is more informative about the range of effect sizes compatible with the data.

The Phase 2 counseling comparison and the chronic-pain comparisons also have confidence intervals spanning zero. Their p-values are 0.348, 0.107, and 0.219. These estimates are not precise enough to rule out potentially meaningful differences in either direction. The registry record does not state what size of difference would be clinically important, so statistical uncertainty cannot be translated into a conclusion about clinical equivalence.

Precision depends on the number of observed outcomes and the number of participants contributing information, as well as on the analysis method and outcome variability. The primary analysis includes the full Phase 1 population stated in the registry, but the successful outcome counts are relatively small compared with the arm sizes. That pattern makes it important not to over-read the direction suggested by the counts. A confidence interval spanning zero is not proof of no effect; it indicates that the available data do not sharply distinguish among a range of possibilities.

Confidence intervals also do not account for every uncertainty. The registry record does not report methods for handling missing data, adjustment for multiple comparisons, or details of outcome assessment. Those omissions limit how fully the precision and robustness of the estimates can be assessed.

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 participants and not total events. The planning illustration uses a two-sided significance level of 0.05, equal allocation between the two arms, and a normal approximation. It sets the effect to detect at the stated fraction of the observed effect. The registry record does not report the trial’s original sample-size assumptions; these figures are illustrations based on the observed results, not a reconstruction of the trial’s design.

The 80% power column describes the number of participants per arm the planning calculation estimates would be needed to have an 80% chance of detecting the selected effect under the stated assumptions. The 90% power column gives the corresponding planning figure for a 90% chance. For the primary comparison, the illustration gives 3,627 participants per arm for 80% power and 4,854 participants per arm for 90% power when planning for 100% of the observed effect. For 75% of the observed effect, the corresponding figures are 6,262 and 8,382 per arm; for 50%, they are 13,669 and 18,299 per arm.

Smaller effects require more participants to detect under the same assumptions, because a smaller difference is harder to distinguish from random variation. The 80% and 90% values are planning probabilities for a hypothetical future study with the specified assumptions; they are not observed power for POATS. They do not change the interpretation of the observed trial result, and they do not show that a future study will necessarily achieve the stated power if its assumptions differ.

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 = 0.058; treatment proportion p₁ = 0.074; average p̄ = 0.066.
2. First term: z_1−α/2 × √[2p̄(1−p̄)] = 1.96 × 0.3509 = 0.6878.
3. Second term: z_1−β × √[p₀(1−p₀) + p₁(1−p₁)] = 0.842 × 0.3507 = 0.2952.
4. n = (0.6878 + 0.2952)² / (0.0163)² = 3,626.1, rounded up to 3,627 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 <- 0.0577507598784195                          # control proportion
p1 <- 0.0740740740740741                          # 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(316277)
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 = 3,627 (80% power), 4,854 (90% power)
 75% of observed effect: n per arm = 6,262 (80% power), 8,382 (90% power)
 50% of observed effect: n per arm = 13,669 (80% power), 18,299 (90% power)
power.prop.test: 3626.1 
Simulated power at n = 3,627 per arm: 0.807 (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 7 outcome measures and 0 analyses in its analysis listing. It presents a primary outcome and selected secondary outcomes, but does not report a multiplicity procedure, such as a correction to control the overall false-positive risk across several tests. When multiple outcomes, timepoints, or subgroup comparisons are considered, some differences can arise by chance even if no underlying effects exist. Unadjusted secondary p-values should therefore be interpreted cautiously, particularly when no prespecified testing hierarchy is reported.

The primary outcome definition permits no more than 1 missing urine sample during the 12 weeks, and success also requires completing week 12. This is part of the endpoint definition; it is not a complete account of how all missing observations were handled in analysis. The registry record does not report imputation, assumptions about missingness, sensitivity analyses, or the reasons for missing data. It also does not state whether the analysis followed an intention-to-treat principle beyond describing the primary analysis population as the 653 participants randomized to Phase 1 and included in analysis.

The record does not report the original statistical analysis plan, the handling of protocol deviations, or whether outcomes were adjudicated without knowledge of counseling condition. These details matter because outcome classification can depend on participant reports, urine test availability, treatment completion, and the stated rules for additional treatment. The registry record gives the success criteria but not the operational detail needed to assess measurement consistency.

For Phase 2, the reported comparison includes 180 participants in each counseling condition. The record does not explain the pathway into this later analysis population or provide a full account of missing follow-up. The Phase 2 estimate therefore applies to the reported analysis group and should not automatically be generalized to all participants randomized at the outset. Taken together, missing-data and multiplicity methods are not reported in the registry record, limiting assessment of the robustness of the secondary findings and some aspects of the primary analysis.

15. 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 5% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: suicidal ideation (0.6%), anxiety (0.3%), ovarian cyst (0.3%), ovarian torsion (0.3%), dehydration (0.3%).

The registry reports serious adverse events for 6 participants in the EMM group, among 329 at risk, and 5 participants in the SMM group, among 324 at risk. These are participant counts, not event rates calculated for comparison. The listed serious-event terms include Suicidal Ideation (0.6%), Anxiety (0.3%), Ovarian Cyst (0.3%), Ovarian Torsion (0.3%), Dehydration (0.3%), and Gastric Ulcer (0.3%). The registry record does not describe the relationship of each event to treatment in this summary.

Other adverse events are listed for 277 participants in EMM and 265 in SMM. The record does not provide a comparative safety analysis or confidence intervals for these counts. Deaths are not reported in the registry record. Differences in raw counts should not be interpreted as proof of a safety difference, especially without information about event definitions, follow-up completeness, severity beyond the serious-event classification, or statistical analysis. These figures describe what the registry reports and do not, by themselves, establish causation.

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

The central result is that the posted Phase 1 counts do not show a clear difference in successful opioid-use outcome between buprenorphine/naloxone with EMM and buprenorphine/naloxone with SMM. The unadjusted comparison has p = 0.4, and its confidence interval spans zero. This means the data do not provide a precise estimate that favors either counseling condition. It does not demonstrate that the conditions are identical or that any possible difference is negligible.

The result applies to the registry’s specific success definition and 12-week Phase 1 timeframe. A participant had to meet conditions involving treatment completion, self-reported use, urine test results, additional substance use disorder treatment, and missing urine samples. Findings for that composite endpoint cannot be assumed to describe every aspect of treatment, longer-term outcomes, or other definitions of success.

The study compared two management conditions alongside buprenorphine/naloxone. It did not compare buprenorphine/naloxone with no treatment, and the registry record does not report enough detail to isolate the effects of individual components within EMM or SMM. Its Phase 2 and chronic-pain analyses ask different questions and are secondary. Those results are also uncertain and do not supply a firm basis for broad conclusions about subgroup effects.

Interpretation is further limited by details not reported in the registry record, including the original sample-size justification, statistical analysis plan, missing-data methods, and multiplicity handling. The adverse-event counts are descriptive and do not establish comparative safety or causation. Taken together, the posted results support a cautious account of the observed outcomes in this trial, not a universal conclusion about treatment effectiveness, equivalence, or safety.

17. Why This Trial Matters Statistically

This trial is statistically informative because its primary outcome is defined through several measurable conditions rather than a single report of opioid use. That makes the result interpretable as a specific composite classification, while also reminding readers that each component contributes to who is counted as successful. The registry’s definition sets clear thresholds for self-report, urine results, additional treatment, completion, and missing samples.

The randomized parallel-group design makes a comparison between EMM and SMM relevant to the study’s stated counseling question. At the same time, the open-label design and absent arm details in the registry record leave questions about the implementation and measurement process unanswered. Statistical interpretation requires attention both to the randomization and to what the record does not explain about outcome ascertainment and analysis.

The primary estimate is not precise enough to support a firm directional conclusion. This makes it a useful example of why a p-value alone is insufficient: the confidence interval communicates that the data remain compatible with a range of differences. The planning illustrations also demonstrate how much larger a hypothetical study could need to be to detect the observed effect under their stated assumptions, especially when planning for a smaller effect. Those figures are not evidence about power achieved in the completed trial.

Finally, the registry contains secondary comparisons and serious adverse event counts, but they do not replace the primary question. Their varied directions, uncertainty, and lack of reported multiplicity adjustment reinforce the value of distinguishing confirmatory results from exploratory patterns. The main statistical lesson is to align each interpretation with its endpoint, population, timeframe, and uncertainty rather than treating all posted contrasts as interchangeable 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.

Several trial-specific limitations affect interpretation of the posted results. The study was open-label, with masking listed as none. The registry record does not describe how outcome measurement was protected from knowledge of treatment assignment, which matters for an endpoint that includes self-reported opioid use and completion criteria. The definitions are stated, but implementation and adjudication details are not reported.

The primary endpoint is a composite requiring multiple conditions. The posted count does not show how many participants failed each component, so it is not possible to identify which parts of the definition most influenced the outcome. The record also does not report missing-data methods, sensitivity analyses, or a detailed account of follow-up completeness. Although the primary analysis population is described, the analytic handling of all randomized participants cannot be fully assessed from the registry record.

The registry does not report the original sample-size justification or statistical analysis plan. It also does not report adjustment for multiplicity across the primary and secondary outcomes. The available secondary analyses therefore warrant caution, and the absence of reported adjustment makes it difficult to distinguish planned confirmatory questions from exploratory comparisons. The chronic-pain comparisons are subgroup descriptions, not randomized comparisons of pain status; no adjusted analysis or interaction test is reported.

Further limitations include limited detail about the content of EMM and SMM, the randomization procedures, and the pathway into the Phase 2 analysis population. The study summary describes long-term follow-up assessments, but results for those assessments are not reported in the posted outcome summaries. Serious adverse event counts are available, but deaths are not reported and no comparative safety analysis is given. These gaps constrain conclusions about long-term outcomes, subgroup effects, and safety.

19. Related Tutorials

20. Related Statistical Calculators

21. Sources

• ClinicalTrials.gov: Prescription Opioid Addiction Treatment Study (POATS) — NCT00316277.
• Registry record last updated 2013-02-06; status last verified 2013-02.
• Results first posted 2012-07-27.
• 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

POATS was a completed Phase 3 randomized, open-label, parallel-group trial sponsored by Mclean Hospital. It enrolled 653 participants dependent on prescription opioid analgesics and compared buprenorphine/naloxone with EMM against buprenorphine/naloxone with SMM. The primary endpoint counted participants attaining the defined successful opioid-use outcome at the end of Phase 1, at 12 weeks.

The primary counts were 19 in EMM and 24 in SMM; the unadjusted p-value was 0.4, and the confidence interval for the risk difference spanned zero. The posted Phase 2 and chronic-pain comparisons also had confidence intervals spanning zero. The registry record does not report the original sample-size assumptions, detailed analysis plan, missing-data methods, or multiplicity adjustment. Serious adverse event counts were 6 in EMM and 5 in SMM; deaths are not reported in the registry record.