Effect modification, interaction and mediation: an overview of theoretical insights for clinical investigators

Figures & data

Table

Type of assessment	Aim of the assessment
Effect modification	Separate exposure effects according to another variable^Citation12,Citation41
Interaction	Evaluate individual and joint effects of exposures^Citation4
Mediation	Evaluate direct and indirect effects of exposures^Citation22,Citation45

Main motivation for the assessment of effect modification, interaction and mediation.

Figure 1 Mean (95% CI) heavy drinking days per week by topiramate (green circle) or placebo (red square) intervention group.

Notes: (A) Main analysis; (B) analysis in three different alleles of the rs2832407 genotype. Adapted from Kranzler et al¹⁴ with permission from the American Journal of Psychiatry (Copyright ©2014). American Psychiatric Association. All Rights Reserved.

Table 1 Relative risk of endometrial cancer in postmenopausal women continuously using combined hormone replacement therapy (HRT), and the effect modification by body mass index (BMI)

Exposure status	Relative risk of endometrial cancer (95% CI)
	BMI <25 kg/m^2	BMI 25–29 kg/m^2	BMI ≥30 kg/m^2
Continuous, combined HRT user	1.07 (0.73–1.56)	0.88 (0.60–1.30)	0.28 (0.14–0.55)
Never HRT user	Reference group	Reference group	Reference group

Note: Data from the Million Women Study.¹⁵

Table 2 Incidence rate of ischemic heart disease after the use of single or combined antithrombotic therapy with low-dose aspirin and warfarin

Allocated treatment group	Incidence rate/1,000 person-years
Double placebo	13.3
Aspirin alone	10.2
Warfarin alone	10.3
Warfarin and aspirin	8.7

Note: Data from the Medical Research Council’s General Practice Research Framework.¹⁷

Table 3 Use of single and combined antithrombotic therapy and risk of serious upper gastrointestinal bleeding in a case–control study using research databases in Denmark, 2000–2004

Exposure group	Odds ratio (95% CI)
No antithrombotic use at all	Reference category
Aspirin use alone	2.4 (2.0–2.8)
Clopidogrel use alone	3.1 (1.7–5.6)
Aspirin and clopidogrel use	12.6 (6.6–24)

Note: Data from Hallas et al.¹⁸

Figure 2 Relation between the exposure – the intermediate or mediator – and the outcome.

Table 4 Total effect of fibromyalgia treatment (pregabalin vs placebo) on restorative sleep improvement (measured using the Sleep Quality Daily Diary score), and indirect effect via pain scale reduction

Type of effect	Mean change in Sleep Quality score
Total effect (irrespective of whether pain reduction occurs)
Pregabalin treatment → restorative sleep	−1.295
Pregabalin treatment → pain reduction → restorative sleep
Indirect or mediated effect (via pain reduction)
Pregabalin treatment → pain reduction → restorative sleep	−0.718

Note: Data from the randomized controlled trial (RCT) published by Mease et al²⁶ and analyzed by Russell et al.²⁷

Table 5 Total effect of blood group (non-O vs O) on the occurrence of deep venous thrombosis, and the direct effect of blood group on the occurrence of deep venous thrombosis, from the case–control Leiden Thrombophilia study

Type of effect	Odds ratio (95% CI)
Total effect (irrespective of clotting factor VIII)
Non-O blood type → deep venous thrombosis	2.0 (1.4–2.8)
Non-O blood type → factor VIII → deep venous thrombosis
Direct effect (adjusting for clotting factor VIII)
Non-O blood type → deep venous thrombosis	1.5 (1.0–2.1)

Note: Data from Le Cessie et al.²⁹

RothmanKJGreenlandSLashTLModern Epidemiology3rd edPhiladelphiaLippincott Williams & Wilkins2008

 Google Scholar

VanderWeeleTJRobinsJMFour types of effect modification: a classification based on directed acyclic graphsEpidemiology200718556156817700242

 PubMed Web of Science ®Google Scholar

RothmanKJSynergy and antagonism in cause-effect relationshipsAm J Epidemiol19749963853884841816

 PubMed Web of Science ®Google Scholar

HafemanDMSchwartzSOpening the Black Box: a motivation for the assessment of mediationInt J Epidemiol200938383884519261660

 PubMed Web of Science ®Google Scholar

RobinsJMGreenlandSIdentifiability and exchangeability for direct and indirect effectsEpidemiology1992321431551576220

 PubMed Web of Science ®Google Scholar

KranzlerHRCovaultJFeinnRTopiramate treatment for heavy drinkers: moderation by a GRIK1 polymorphismAm J Psychiatry2014171444545224525690

 PubMed Web of Science ®Google Scholar

BeralVBullDReevesGCollaboratorsMWSEndometrial cancer and hormone-replacement therapy in the Million Women StudyLancet200536594701543155115866308

 PubMed Web of Science ®Google Scholar

The Medical Research Council’s General Practice Research FrameworkThrombosis prevention trial: randomised trial of low-intensity oral anticoagulation with warfarin and low-dose aspirin in the primary prevention of ischaemic heart disease in men at increased riskLancet199835190982332419457092

 PubMed Web of Science ®Google Scholar

HallasJDallMAndriesAUse of single and combined anti-thrombotic therapy and risk of serious upper gastrointestinal bleeding: population based case-control studyBMJ2006333757172616984924

 PubMed Web of Science ®Google Scholar

MeasePJRussellIJArnoldLMA randomized, double-blind, placebo-controlled, phase III trial of pregabalin in the treatment of patients with fibromyalgiaJ Rheumatol200835350251418278830

 PubMed Web of Science ®Google Scholar

RussellIJCroffordLJLeonTThe effects of pregabalin on sleep disturbance symptoms among individuals with fibromyalgia syndromeSleep Med200910660461019410509

 PubMed Web of Science ®Google Scholar

Le CessieSDebeijJRosendaalFRCannegieterSCVandenbrouckeJPQuantification of bias in direct effects estimates due to different types of measurement error in the mediatorEpidemiology201223455156022526092

 PubMed Web of Science ®Google Scholar