This page provides an independent statistical analysis and educational interpretation of publicly reported results. ClinicalTrials.gov provides the official trial registry record. Numerical trial results on this page are restricted to the results posted in the ClinicalTrials.gov record; calculations added by Clinical Biostats are labelled as such.
1. Trial at a Glance
This page covers the registry results for Lenalidomide in Older Patients With Acute Myeloid Leukemia Without Chromosome 5q Abnormalities (NCT00546897), a non-randomized phase 2 study. The primary outcome is Complete Remission Rate (CRm + CRi + CRc).
| Feature | Registry value |
|---|---|
| ClinicalTrials.gov identifier | NCT00546897 |
| Phase | Phase 2 |
| Condition | Leukemia, Myeloid, Acute |
| Allocation | Non-Randomized |
| Design model | Parallel |
| Masking | None (open label) |
| Enrollment | 48 |
| Arms in the primary outcome | n/a |
| Lead sponsor | Washington University School of Medicine |
| Results posted | Yes |
2. Clinical Question
Population
Older participants with untreated acute myeloid leukemia without chromosome 5q abnormalities. The official title specifies an age of at least 60 years. The registry record reports enrollment of 48 participants.
Intervention
Lenalidomide, given orally in two non-randomized cohorts with different treatment schedules. Cohort 1 began with high-dose treatment and could proceed to low-dose treatment if there was no progressive disease. Cohort 2 began with high-dose induction, with later treatment dependent on response.
Comparator
There was no randomized control group. The cohorts are not a like-for-like control comparison, so differences between them cannot be attributed to treatment schedule alone.
Primary question
What number of participants met the registry-defined complete remission endpoint, combining CRm, CRi, and CRc, after two cycles of low-dose lenalidomide?
The study was designed to examine the safety and efficacy of lenalidomide in older participants with untreated acute myeloid leukemia without chromosome 5q abnormalities. Its primary endpoint is a remission count at a specified treatment stage, rather than a direct comparison with another treatment or with no treatment.
The endpoint’s timing is described as approximately Day 113 for Cohort 1 and approximately Day 104 for Cohort 2, after two low-dose cycles. Those timings differ by cohort. In addition, the registry notes that 9 out of 15 participants in Cohort 1 and 23 out of 33 participants in Cohort 2 did not receive the two low-dose cycles. The reasons for not receiving those cycles are not reported in the registry record.
These details matter when interpreting the question. The endpoint concerns a remission classification at a later treatment stage, while a substantial portion of each cohort did not receive the specified low-dose treatment. The registry record does not report how the primary endpoint was summarized for participants who did not reach that stage, nor does it report the remission count itself. It therefore does not establish the observed primary endpoint result or permit a treatment-effect estimate against a control.
3. Trial Design
This was a completed Phase II study sponsored by Washington University School of Medicine. The registry record gives a start date of 2007-02 and a primary completion date of 2010-06. It describes a non-randomized, parallel-cohort design with no masking. Enrollment was 48 participants with acute myeloid leukemia.
Participants were assigned to Cohort 1 or Cohort 2, but the registry record does not report a random allocation process. The cohorts therefore should not be treated as randomized arms. Any differences in outcomes between them could reflect differences in the participants, treatment schedules, care, or other factors, rather than the schedule alone. The record itself cautions that these are not a like-for-like control comparison.
The Cohort 1 description begins with oral lenalidomide at 50 mg/day for 14 days followed by 30 days of rest, and also describes 50 mg/day for 21 days as Cycle 1 and Cycle 2. If there was no progressive disease, the description then specifies 10 mg/day for 28 days for 12 cycles. Cohort 2 is described as receiving oral lenalidomide 50 mg/day for 28 days as Cycle 1 induction. Subsequent treatment depended on response: participants obtaining complete remission would proceed to low-dose treatment, while participants with a non-CR would receive a second high-dose cycle. Further detail in the Cohort 2 description is not reported in the registry record.
The primary endpoint was assessed after two cycles of low-dose lenalidomide, at approximately Day 113 for Cohort 1 and approximately Day 104 for Cohort 2. The registry reports that 9 out of 15 participants in Cohort 1 and 23 out of 33 participants in Cohort 2 did not receive the two low-dose cycles. It does not report the reasons or the endpoint-handling approach for those participants. A protocol, sample-size justification, interim-analysis plan, and detailed statistical analysis plan 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
| Stage | Cohort 1 | Cohort 2 |
|---|---|---|
| STARTED | 15 | 33 |
| COMPLETED | 15 | 33 |
| NOT COMPLETED | 0 | 0 |
Analysis population for the primary outcome, as defined in the registry: 9 out of 15 participants did not receive the two low dose cycles of lenalidomide in Cohort 1. 23 out of 33 participants did not receive the two low dose cycles of lenalidomide in Cohort 2.
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.
| Characteristic | Unit | Cohort 1 | Cohort 2 | Total |
|---|---|---|---|---|
| Participants analysed | 15 | 33 | 48 | |
| Age, Continuous | years | 71 (60 to 86) | 71 (60 to 88) | 71 (60 to 88) |
| Sex: Female, Male: Female | Participants | 4 | 12 | 16 |
| Sex: Female, Male: Male | Participants | 11 | 21 | 32 |
| Region of Enrollment: United States | participants | 15 | 33 | 48 |
| Eastern Cooperative Oncology Group (ECOG) Performance Status: 0 | participants | 10 | 16 | 26 |
| Eastern Cooperative Oncology Group (ECOG) Performance Status: 1 | participants | 4 | 13 | 17 |
| Eastern Cooperative Oncology Group (ECOG) Performance Status: 2 | participants | 1 | 4 | 5 |
| Cytogenetic Risk Category: Intermediate | participants | 13 | 18 | 31 |
| Cytogenetic Risk Category: Unfavorable | participants | 2 | 13 | 15 |
| Cytogenetic Risk Category: Unknown | participants | 0 | 2 | 2 |
6. Primary Endpoint
The registry lists the primary outcome as Complete Remission Rate (CRm + CRi + CRc), measured over After 2 cycles of low dose lenalidomide (approximately Day 113 for Cohort 1 and approximately Day 104 for Cohort 2).
| Element | Registry entry |
|---|---|
| Result type | Number |
| Measure of spread | not applicable |
| Unit | participants |
| Endpoint classification | Binary Endpoint |
The primary endpoint is the number of participants meeting a combined complete remission definition: CRm, CRi, or CRc. The registry describes the endpoint as assessed after two cycles of low-dose lenalidomide, approximately Day 113 for Cohort 1 and approximately Day 104 for Cohort 2. The posted primary result is not reported in the registry record.
CRm is defined in the record as morphologic leukemia-free state, with fewer than 5% blasts in a bone marrow aspirate with marrow spicules and a count of more than 200 nucleated cells, no blasts with Auer rods, and no persistent extramedullary disease. The description also specifies neutrophil and platelet count thresholds and independence from transfusions for a minimum of 1 week before each marrow assessment. It states that there is no duration requirement for this designation.
CRi is described as complete remission with the exception of neutropenia below 1000/uL or thrombocytopenia below 100,000/ul. CRc applies only to participants with an identified cytogenetic abnormality and requires morphologic complete remission plus reversion to a normal karyotype, with no clonal abnormalities detected in a minimum of 20 mitotic cells.
In general, combining remission categories can capture participants who satisfy different specified forms of response. It is important, however, to interpret a combined endpoint according to its exact definitions and assessment timing. The record does not report how many participants met each component, how the components were combined in the analysis, or how participants who did not reach the low-dose cycles were handled.
7. Why No Re-analysis Is Shown
8. Statistical Interpretation
The registry record identifies a primary endpoint and provides its definitions and assessment timing, but it does not report the number of participants who achieved the combined remission endpoint. Consequently, the primary remission result cannot be described from the registry information available for this analysis. The record also lists no statistical analyses. It therefore does not provide an estimate of the endpoint’s precision, such as a confidence interval, or a statistical test for the primary outcome.
The study’s non-randomized design is central to interpretation. Cohort 1 and Cohort 2 followed different treatment schedules and were not assigned through a reported random process. They are not equivalent to an intervention group and a control group. A numerical difference between cohorts, if reported elsewhere, would not by itself establish that one schedule caused a better or worse outcome. Differences in participant characteristics, response-dependent treatment, or progression could also be relevant, but those factors are not described sufficiently in the registry record to assess them.
Interpretation of the primary endpoint is also affected by treatment exposure. The record says that 9 out of 15 participants in Cohort 1 and 23 out of 33 participants in Cohort 2 did not receive the two low-dose cycles that define the endpoint’s assessment stage. This indicates that many participants did not reach the specified treatment stage. The record does not state why they did not receive those cycles, whether they were assessed for remission at another time, or how their outcomes entered the primary endpoint analysis. It would be inappropriate to assume that they were all treatment failures, were all excluded, or were all missing at random.
The safety information reports serious adverse events affecting 10 participants in Cohort 1 and 25 in Cohort 2. It also reports other adverse events affecting 15 participants in Cohort 1 and 33 in Cohort 2. These are descriptive counts, not a randomized comparison. The listed serious-event terms include febrile neutropenia, pneumonia, and death due to progressive disease, among others. The registry’s term-specific percentages do not establish whether an event was caused by lenalidomide.
Overall, the record describes the study design, treatment schedules, endpoint definitions, and selected safety summaries. It does not provide the primary remission count or a formal statistical analysis. The evidence presented in the registry record therefore supports a description of the trial and its reported events, but not a quantified estimate of the remission outcome or a causal comparison between the cohorts.
9. Statistical Methods Explained
What does a non-randomized design imply?
In general, randomization is used to reduce systematic differences between treatment groups at the time of assignment. This study is explicitly non-randomized, and the registry record describes two cohorts with different lenalidomide schedules. The cohorts are therefore not a like-for-like control comparison. Any contrast between them would be descriptive unless supported by an appropriate analysis addressing differences between participants and treatment pathways. The registry record does not report such an adjusted comparison.
The treatment pathways also depended on disease progression or response. In Cohort 1, the record says that participants without progressive disease could proceed to low-dose treatment. In Cohort 2, later treatment depended on the response to induction. This means that treatment received after the initial course was not described as identical for all participants. In general, when subsequent treatment depends on response, outcomes in later treatment stages can reflect both the treatment pathway and the characteristics of those who reached that stage.
What does the remission endpoint capture?
The endpoint is a participant count combining CRm, CRi, and CRc. These categories have specific definitions in the registry record. CRm is described using bone marrow findings, disease features, blood-count thresholds, and transfusion independence. CRi allows an exception for neutropenia or thrombocytopenia. CRc is restricted to participants with an identified cytogenetic abnormality and requires a normal karyotype under the stated definition.
In general, a composite endpoint combines multiple outcomes into one summary. It can represent more than one clinically defined response category, but its interpretation depends on knowing which components occurred and how the combined result was calculated. The registry record does not report the component counts or the overall remission count. It also does not report whether participants could meet more than one component or how overlapping classifications were handled.
The endpoint is timed after two low-dose cycles, with different approximate assessment days stated for the cohorts. A time point tied to treatment exposure is not necessarily the same for participants who stop or change treatment earlier. The record notes that 9 out of 15 participants in Cohort 1 and 23 out of 33 participants in Cohort 2 did not receive those two cycles. It does not explain how assessments for these participants were handled. In general, an endpoint tied to reaching a treatment stage requires transparent reporting of the analysis population and the handling of participants who do not reach that stage.
How should counts and precision be understood?
The primary endpoint is expressed as a number of participants. A count describes how many participants met a definition, but a count alone does not show how uncertain the underlying response proportion may be. In general, a confidence interval gives a range of values compatible with the data and the chosen statistical method. Its width reflects statistical precision; it does not correct for non-random assignment, incomplete follow-up, or other design limitations.
No primary endpoint count, confidence interval, or hypothesis test is reported in the registry record. No analysis is listed for the primary outcome. As a result, no inferential conclusion about remission can be drawn from a significance test, and no precision estimate can be quoted. The absence of a reported analysis is distinct from evidence that the endpoint had no effect or that groups were equivalent.
The record describes planning illustrations based on a two-sided significance level of 0.05, equal allocation between arms, and a normal approximation. Such illustrations are hypothetical calculations for planning under stated assumptions; they are not observed trial results or observed power. The registry’s stated assumptions also say that the effect to detect is a fraction of an observed effect. No planning values need to be interpreted as evidence about what this study observed.
Why does over-dispersion matter in general?
Over-dispersion is a general statistical issue that arises when observed variability is greater than a model expects. If it is ignored, standard errors can be too small and confidence intervals can be too narrow, making results appear more precise than warranted. The registry record does not report a model for the primary endpoint or whether over-dispersion was assessed. This concept is therefore relevant as general statistical context, not as a feature established for this trial.
What can safety summaries establish?
The registry lists participants affected by serious and other adverse events in each cohort, along with selected serious-event terms and percentages. In general, descriptive safety counts summarize recorded events or affected participants according to the registry’s reporting convention. They do not by themselves establish that the treatment caused an event, compare risks under randomized conditions, or show whether an event was more common than it would have been without treatment. The record does not report an inferential safety comparison or event-attribution analysis.
Interpretation of safety percentages depends on their denominator, event definitions, follow-up period, and whether a participant could contribute more than one event. The registry’s adverse-event summary identifies cohort-specific counts at risk and affected, but further details such as the full event-level accounting and statistical analysis are not reported in the registry record. Those limitations should be kept separate from the primary remission endpoint, which is not a safety measure.
10. Multiplicity, Missing Data and Analysis Populations
The registry record lists multiple secondary outcomes, including safety and tolerability, response rate, morphologic leukemia-free state, several remission categories, partial remission, and overall survival. It does not report a statistical plan describing how these outcomes were prioritized beyond identifying the primary endpoint, or how multiple statistical tests were to be handled. A multiplicity adjustment or a prespecified hierarchy is not reported in the registry record.
In general, testing many outcomes can increase the chance of obtaining a seemingly notable result by chance. Methods such as a prespecified hierarchy or adjustment of significance thresholds can address that concern, but the registry does not state whether any such method was planned or used. Because no analyses are listed, no conclusions about multiplicity-adjusted or unadjusted statistical significance can be made.
The registry reports that 9 out of 15 participants in Cohort 1 and 23 out of 33 participants in Cohort 2 did not receive the two low-dose cycles specified before the primary endpoint assessment. This raises an important question about the analysis population and missing endpoint information. The record does not report whether these participants were included in the primary analysis, assessed at another time, treated as not achieving remission, or excluded. It also does not report reasons for not receiving the cycles, the completeness of outcome collection, or any method for handling missing data.
The registry record does not describe an intention-to-treat population, a per-protocol population, or another analysis-population definition. It also does not report sensitivity analyses for missing or unevaluable outcomes. In general, these details are needed to understand how the analyzed participants relate to those enrolled and how robust the result is to assumptions about unavailable outcomes. Since the primary endpoint count and analysis are not reported, the effect of these issues on the remission estimate cannot be quantified.
11. Safety Results
The registry reports serious adverse events as affected participants divided by participants at risk.
| Arm | Participants with serious adverse events | Participants at risk | Percentage |
|---|---|---|---|
| Cohort 1 | 10 | 15 | 66.7% |
| Cohort 2 | 25 | 33 | 75.8% |
Other (non-serious) adverse events
Participants with at least one non-serious event at or above the registry's 0% reporting threshold.
| Arm | Participants affected of at risk |
|---|---|
| Cohort 1 | 15 of 15 |
| Cohort 2 | 33 of 33 |
Most frequent serious adverse event terms
| Event term | Organ system | Cohort 1 | Cohort 2 |
|---|---|---|---|
| Febrile neutropenia | Blood and lymphatic system disorders | 6 of 15 | 12 of 33 |
| Pneumonia | Infections and infestations | 2 of 15 | 9 of 33 |
| Death due to progressive disease | General disorders | 3 of 15 | 6 of 33 |
| Vomiting | Gastrointestinal disorders | 1 of 15 | 2 of 33 |
| Dehydration | Gastrointestinal disorders | 1 of 15 | 2 of 33 |
| Lower extremity weakness | Musculoskeletal and connective tissue disorders | 1 of 15 | 2 of 33 |
| Fatigue | General disorders | 0 of 15 | 3 of 33 |
| Syncope | Cardiac disorders | 0 of 15 | 3 of 33 |
| Anorexia | Metabolism and nutrition disorders | 0 of 15 | 3 of 33 |
| Nausea | Gastrointestinal disorders | 1 of 15 | 1 of 33 |
Most frequent non-serious adverse event terms
| Event term | Organ system | Cohort 1 | Cohort 2 |
|---|---|---|---|
| Platelets | Investigations | 12 of 15 | 25 of 33 |
| Hypocalcemia | Investigations | 12 of 15 | 21 of 33 |
| Hemoglobin | Blood and lymphatic system disorders | 10 of 15 | 22 of 33 |
| Hypoalbuminemia | Investigations | 8 of 15 | 19 of 33 |
| SGPT (ALT) | Investigations | 12 of 15 | 14 of 33 |
| Fatigue | General disorders | 8 of 15 | 17 of 33 |
| Leukocytes (WBC) | Investigations | 9 of 15 | 15 of 33 |
| Rash | Skin and subcutaneous tissue disorders | 6 of 15 | 18 of 33 |
| Dyspnea | Respiratory, thoracic and mediastinal disorders | 8 of 15 | 13 of 33 |
| SGOT (AST) | Investigations | 8 of 15 | 12 of 33 |
Serious events with the highest arm-level rate: febrile neutropenia (40.0%), pneumonia (27.3%), death due to progressive disease (20.0%), fatigue (9.1%), syncope (9.1%).
The registry reports serious adverse events for Cohort 1, with 10 participants affected among 15 at risk, and for Cohort 2, with 25 participants affected among 33 at risk. For other adverse events, it reports 15 affected among 15 at risk in Cohort 1 and 33 affected among 33 at risk in Cohort 2. These are descriptive cohort summaries. The cohorts were not randomized against one another, so the counts should not be read as a comparative treatment effect.
The listed serious-event terms and percentages are febrile neutropenia, 40.0%; pneumonia, 27.3%; death due to progressive disease, 20.0%; fatigue, 9.1%; syncope, 9.1%; and anorexia, 9.1%. These are reported as serious-event terms in the registry. The record does not provide enough detail to determine whether the percentages refer to participants with each event under a common follow-up definition, whether participants could appear in multiple terms, or how causality was assessed.
In general, an adverse event occurring during a study is not automatically caused by the study treatment. Descriptive event counts also do not establish the comparative risk that would have occurred under another treatment or without treatment. The registry record does not report a formal safety comparison, attribution analysis, or detailed time-at-risk analysis. The figures describe what was recorded, not a causal conclusion.
12. What the Result Does, and Does Not, Mean
This trial record describes a Phase II, non-randomized study of lenalidomide in older participants with untreated acute myeloid leukemia without chromosome 5q abnormalities. It specifies two treatment cohorts, a primary remission endpoint combining CRm, CRi, and CRc, and selected adverse-event summaries. It also reports that many participants did not receive the two low-dose cycles that precede the stated primary endpoint assessment.
The record does not report the number of participants who achieved the primary endpoint. It therefore does not establish the remission frequency, its precision, or a statistical test result. Without a reported endpoint estimate, the registry record cannot show how the observed remission outcome compares with a target, another treatment, or an expected rate.
The non-randomized design also limits causal interpretation. Cohort 1 and Cohort 2 followed different treatment schedules and are not described as a randomized comparison. Even if cohort-specific outcomes were available, an unadjusted difference would not isolate the effect of the schedule from other differences between the cohorts. The registry record does not report an adjusted treatment comparison.
The safety counts and listed serious-event percentages provide descriptive information about events recorded in the study. They do not establish that lenalidomide caused those events or that the event frequency would be different under another treatment. Event interpretation is further constrained because details on attribution, follow-up, and full event accounting are not reported in the registry record.
In general, a Phase II study can provide descriptive evidence about treatment delivery, response classifications, and safety observations that may inform later research. That general role does not substitute for the missing primary endpoint result or a comparative design. The registry information cannot support a conclusion that the treatment is effective or ineffective, nor can it establish equivalence between the cohorts. It supports only a cautious account of the registered design, definitions, participant treatment exposure, and reported safety summaries.
13. Why This Trial Matters Statistically
This study illustrates why a clearly named endpoint does not by itself provide an interpretable treatment result. The registry defines a combined remission measure and specifies when it should be assessed, but does not report the remission count or a statistical analysis. Without those results, readers cannot evaluate the observed response frequency or its precision.
It also illustrates the importance of distinguishing a treatment cohort from a control group. The study was non-randomized, and the cohorts had different treatment schedules. They are not a like-for-like control comparison. Treating their outcomes as if they came from randomized arms would overstate what the design can establish.
The reported treatment exposure is relevant to the endpoint’s meaning. The primary outcome was timed after two low-dose cycles, while 9 out of 15 participants in Cohort 1 and 23 out of 33 participants in Cohort 2 did not receive those cycles. Without the reasons and analysis handling, readers cannot determine whether the reported endpoint—if available—would represent everyone enrolled, only participants reaching that stage, or another population.
For biostatisticians and clinical researchers, the registry record highlights several reporting needs: the primary outcome count and precision, analysis-population definitions, handling of participants who did not reach the assessment stage, and a prespecified approach to multiple outcomes. Those are general standards for interpreting a clinical trial record; whether any particular methods were used in this study is not reported in the registry record.
14. Limitations
- The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
- Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.
The primary limitation is that the registry record does not report the primary remission count or its statistical analysis. As a result, neither the magnitude nor the precision of the primary endpoint can be evaluated. The record also does not report a confidence interval, hypothesis test, or adjusted estimate.
The non-randomized design prevents a straightforward causal comparison between Cohort 1 and Cohort 2. Their treatment schedules differed, and the record does not describe random assignment or a statistical adjustment for cohort differences. Any cohort contrast should therefore be treated as descriptive, not as a randomized estimate of treatment effect.
The endpoint is assessed after two low-dose cycles, but the record says that 9 out of 15 participants in Cohort 1 and 23 out of 33 participants in Cohort 2 did not receive those cycles. Reasons for this treatment exposure pattern and the endpoint-handling rules are not reported in the registry record. This limits understanding of the population represented by the endpoint.
The Cohort 2 treatment description is incomplete in the registry record, and the Cohort 1 schedule is described with multiple high-dose timing details. The record does not provide a complete protocol to clarify additional treatment procedures. It also does not report a detailed statistical analysis plan, multiplicity approach, missing-data method, or analysis-population definition.
Safety findings are limited to descriptive counts and selected serious-event terms and percentages. A detailed account of follow-up, event attribution, and event-level analysis is not reported in the registry record. These limitations constrain interpretation of both efficacy and safety; they do not establish that an unreported analysis was or was not performed outside the registry record.
15. Related Tutorials
16. Related Statistical Calculators
17. Sources
- ClinicalTrials.gov: Lenalidomide in Older Patients With Acute Myeloid Leukemia Without Chromosome 5q Abnormalities — NCT00546897.
- Registry record last updated 2014-09-29; status last verified 2014-09.
- Results first posted 2014-09-29.
- Data retrieved from ClinicalTrials.gov on 2026-11-09.
- Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.
Continue through the Clinical Biostats statistical pathway
Use this trial's endpoint as a starting point for deeper study of effect measures, confidence intervals, and sample-size planning.
18. Record Summary
The registry record describes a completed, non-randomized Phase II study sponsored by Washington University School of Medicine. The study enrolled 48 participants with acute myeloid leukemia and evaluated oral lenalidomide in two cohorts. The official title specifies older participants, at least 60 years, with untreated disease without chromosome 5q abnormalities.
The primary endpoint is the number of participants meeting a combined complete remission definition of CRm, CRi, or CRc after two low-dose cycles. The record gives approximate assessment timing for each cohort and notes that 9 out of 15 participants in Cohort 1 and 23 out of 33 participants in Cohort 2 did not receive those cycles. The primary endpoint result and its analysis are not reported in the registry record.
Safety summaries report serious and other adverse-event counts by cohort and selected serious-event terms. The record does not report a randomized comparison, detailed analysis plan, missing-data approach, or multiplicity method. The findings are therefore best read as a description of the registered design and recorded safety information, not as a comparative estimate of treatment effect.