Indirect comparisons are undertaken when a comparison is made between two regimens, usually where the regimens have never been given concurrently in any controlled trial investigating the same general patient population. We highlight the issues of making indirect comparisons when there has been a period of time between the studies, particularly when indirect comparison is being made to placebo. We discuss the impact of any bias in indirectly estimating any effect over placebo in context with noninferiority trials.
HasselbladVKongDF. Statistical methods for comparison to placebo in active-control trials. Drug Inf J.2001; 25:435–449.
2.
LimEAliZAliAIndirect comparison meta-analysis of aspirin therapy after coronary surgery. Br Med J.2003; 327:1309–1313.
3.
SongFAltmanDGGlennyAMDeeksJJ. Validity of indirect comparison for estimating efficacy of competing interventions: empirical evidence from published meta-analyses. Br Med J.2003; 326:472.
4.
WangSJHungHMJTsongY.Non-inferiority analysis in active controlled trials. Encyclopaedia of Biopharmaceutical Statistics,2nd ed.New York: Marcel Dekker; 2003:674–677.
5.
D'AgostinoRBMassaroJSullivanLM. Non-inferiority trials: design concepts and issues—the encounters of academic consultants in statistics. Stat Med.2003; 22:169–186.
6.
WangSJHungHMJTsongY.Utility and pitfalls of some statistical methods in active controlled trials. Control Clin Trials.2002; 23:15–28.
7.
SnapinnSM. Alternatives for discounting in the analysis of non-inferiority trials. J Biopharm Stat.2004; 14:263–273.
8.
BechP.Meta analysis of placebo controlled trials with mirtazipine using core items of the Hamilton depression scale as evidence of pure anti-depressive effect in the short term treatment of major depression. Int J Neuropsychopharmacol.2001; 4:337–345.
9.
FurukawaTMcGuireHBarbuiC.Low dosage tricyclic antidepressants for depression. Cochrane Database.2003.
10.
WilsonKMotramPSvanranthanPNightingaleA.Antidepressants versus placebo for the depressed elderly. Cochrane Database.2003.
11.
HazellPO'ConnelDHeathcoteDHenryD.Tricyclic drugs for depression in children and adolescents. Cochrane Database.2003.
12.
WalshBTSneidmanSNSyskoRGouldM.Placebo response in studies of major depression, variable, substantial and growing. JAMA.2002; 14: 1840–1847.
13.
McClungCQuessySJuliousSSegrettiABlumD.Placebo response rates in geographical location in COX-2 inhibitor trials of rheumatoid arthritis (RA) and osteoarthritis (OA). Paper presented at American College of Rheumatology Conference, San Antonio, 2004.
JuliousSA. Tutorial in biostatistics: sample sizes for clinical trials with normal data. Stat Med.2004; 23:1921–1986.
21.
WangSJHungHMJ. TACT method for non-inferiority testing in active controlled trials. Stat Med.2003; 22:227–238.
22.
DattaSHalloranMELonginiIM. Augmented HIV vaccine trial design for estimating reduction in infectiousness and protective efficacy. Stat Med.1998; 17:185–200.
