Abstract
The results of a randomized clinical trial can be reported using relative and/or absolute estimators of treatment effect. These various measures convey different information, and the choice can influence the physician's appreciation of the size of treatment effect and, subsequently, treatment decisions. We compare the estimators with respect to the clinically relevant information conveyed to physicians, and identify which clinical questions can and cannot be answered directly by each. We also identify opportunities for misinterpretation when one estimator is substituted for another, or when an estimator is mislabeled. Clinically important questions are addressed most directly by reporting both relative and absolute effects using relative risk and its complement, relative risk reduction, and risk difference and its reciprocal, number needed to treat. This is true of estimates of treatment effect derived from a single trial and also from meta-analysis of a group of trials. Because the control group's risk affects the numerical value of the odds ratio, the odds ratio cannot substitute for the risk ratio in conveying clinically important information to physicians. This is especially important when large treatment effects are shown in trials carried out in populations at high baseline risk.
Keywords
Relative riskRandomized controlled trialAbsolute risk reductionOdds ratioMedicineEstimatorClinical trialStatisticsOddsComplement (music)Treatment and control groupsNumber needed to treatInternal medicineMathematicsConfidence intervalLogistic regression
MeSH Terms
HumansMeta-Analysis as TopicOutcome AssessmentHealth CareRandomized Controlled Trials as TopicRisk
Affiliated Institutions
Related Publications
Chemotherapy in adult high-grade glioma: a systematic review and meta-analysis of individual patient data from 12 randomised trials
Trials on the effect of systemic chemotherapy on survival and recurrence in adults with high-grade glioma have had inconclusive results. We undertook a systematic review and met...
Overview of results of randomized clinical trials in heart disease. II. Unstable angina, heart failure, primary prevention with aspirin, and risk factor modification
THIS is the second of a two-part article on the results of randomized trials of treatments designed to reduce mortality or major morbidity in heart disease. Our goal is to summa...
Corticosteroids as Adjunctive Therapy in the Treatment of Influenza: An Updated Cochrane Systematic Review and Meta-analysis
Objectives: Corticosteroids may be beneficial in sepsis, but uncertainty remains over their effects in severe influenza. This systematic review updates the current evidence rega...
Effect of Antihypertensive Drug Treatment on Cardiovascular Outcomes in Women and Men
Background: Trials of drug therapy for hypertension have shown that such therapy has a clear overall benefit in preventing cardiovascular disease. Although these trials have inc...
Effects of Adjuvant Tamoxifen and of Cytotoxic Therapy on Mortality in Early Breast Cancer. An Overview of 61 Randomised Trials Among 28,896 Women
We sought information worldwide on mortality according to assigned treatment in all randomised trials that began before 1985 of adjuvant tamoxifen or cytotoxic therapy for early...
Publication Info
- Year
- 1994
- Type
- article
- Volume
- 47
- Issue
- 8
- Pages
- 881-889
- Citations
- 270
- Access
- Closed
External Links
Social Impact
Altmetric
PlumX Metrics
Social media, news, blog, policy document mentions
Citation Metrics
270
OpenAlex
6
Influential
204
CrossRef
Cite This
John C. Sinclair,
Michael B. Bracken
(1994).
Clinically useful measures of effect in binary analyses of randomized trials.
Journal of Clinical Epidemiology
, 47
(8)
, 881-889.
https://doi.org/10.1016/0895-4356(94)90191-0
Identifiers
- DOI
- 10.1016/0895-4356(94)90191-0
- PMID
- 7730891