Sufficient Cause Interactions and Statistical Interactions
The DOI matches its Crossref record; 20 of the 20 references it cites were found in a registry or an index.
Abstract
When the outcome and all exposures of interest are binary it is sometimes possible to draw conclusions from empirical data about mechanistic interactions in the sufficient cause sense. Empirical conditions are given for sufficient cause interactions and these conditions are compared with and contrasted to interaction coefficients in linear, log-linear and logistic regression models. Conditions that suffice to allow for the interpretation of statistical interactions as sufficient cause interactions are derived. Discussion is presented concerning the implications of the inclusion of confounding variables in the model.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Response patterns of this type might be considered a special case of the syndemic model in the left panel of the Figure and are often referred to as a “sufficient cause interaction.”20, 21”
From Co-occurring epidemics, syndemics, and population health (2017) · cites this paper as [20]
In the citing paper
“If both exposures have neutral or causative effects for all individuals (i.e. the effects are monotonic), RERI > 0 implies such synergism;18,20 without monotonicity, one can still test for synergism by testing RERI > 1.19,20”
From Recommendations for presenting analyses of effect modification and interaction (2012) · cites this paper as [20]
Citations
Cites
A Modified Poisson Regression Approach to Prospective Studies with Binary Data
Zou, G. (2004) · American Journal of Epidemiology · doi:10.1093/aje/kwh090 · Cited by 9,615
Easy SAS Calculations for Risk or Prevalence Ratios and Differences
Spiegelman, D. (2005) · American Journal of Epidemiology · doi:10.1093/aje/kwi188 · Cited by 1,738
Confidence interval estimation of interaction
Hosmer, D. W., & Lemeshow, S. (1992) · Epidemiology · doi:10.1097/00001648-199209000-00012 · Cited by 1,239
ROTHMAN, K. J. (1976) · American Journal of Epidemiology · doi:10.1093/oxfordjournals.aje.a112335 · Cited by 776
Greenland, S. (2004) · American Journal of Epidemiology · doi:10.1093/aje/kwh221 · Cited by 727
BINOMIAL REGRESSION IN GLIM: ESTIMATING RISK RATIOS AND RISK DIFFERENCES1
WACHOLDER, S. (1986) · American Journal of Epidemiology · doi:10.1093/oxfordjournals.aje.a114212 · Cited by 623
Knol, M. J., van der Tweel, I., Grobbee, D. E., Numans, M. E., & Geerlings, M. I. (2007) · International Journal of Epidemiology · doi:10.1093/ije/dym157 · Cited by 444
An overview of relations among causal modelling methods
Greenland, S., & Brumback, B. (2002) · International Journal of Epidemiology · doi:10.1093/ije/31.5.1030 · Cited by 406
Confidence intervals for measures of interaction
Assmann, S. F., Hosmer, D. W., Lemeshow, S., & Mundt, K. A. (1996) · Epidemiology · doi:10.1097/00001648-199605000-00012 · Cited by 262
A modified least-squares regression approach to the estimation of risk difference.
Cheung, Y. B. (2007) · American Journal of Epidemiology · doi:10.1093/aje/kwm223 · Cited by 189
Cited by
Items 101–164 of 164
Bounds on sufficient-cause interaction
Sjölander, A., Lee, W., Källberg, H., & Yudi Pawitan, Y. (2014) · European Journal of Epidemiology · doi:10.1007/s10654-014-9953-9 · Cited by 10
“VanderWeele [6] clarified the relation between statistical interaction (i.”
BackgroundTraskin, M., Wang, W., Have, T. R. T., & Small, D. S. (2012) · Biostatistics · doi:10.1093/biostatistics/kxs019 · Cited by 10
“Besides their use in estimating attributable fractions, monotonicity constraints are often used in other contexts such as testing for sufficient cause interactions on the basis of contrasts between regression coefficients (VanderWeele, 2009).”
Hedström, A. K., Hössjer, O., Hillert, J. A., Stridh, P., Kockum, I. S., Olsson, T. P., & Alfredsson, L. S. (2019) · Multiple Sclerosis Journal · doi:10.1177/1352458519877685 · Cited by 10
Combined effect of work from home and work during nonwork time on sleep disturbance
Lim, J.-Y., Lee, H., Park, J. B., Lee, K. J., Jeong, I., & Jung, J. (2023) · Annals of Occupational and Environmental Medicine · doi:10.35371/aoem.2023.35.e28 · Cited by 11
Losada, A. F., Wei, Y. B., Åberg, E., Sjöholm, L. K., Lavebratt, C., & Forsell, Y. (2012) · Journal of Affective Disorders · doi:10.1016/j.jad.2012.06.034 · Cited by 11
“It has been argued that the calculation of statistical interaction on the additive scale is more indicative of the underlying biological mechanism (Ahlbom and Alfredsson 2005; Greenland and Poole, 1988; Koopman and Weed, 1990; Rothman et al., 2008; VanderWeele, 2009; Zou, 2008).”
Testing for Mechanistic Interactions in Long-Term Follow-Up Studies
Lin, J., & Lee, W. (2015) · PLoS ONE · doi:10.1371/journal.pone.0121638 · Cited by 9
Hedström, A. K., Huang, J., Brenner, N., Butt, J., Kockum, I. S., Waterboer, T., Olsson, T. P., & Alfredsson, L. S. (2021) · European Journal of Neurology · doi:10.1111/ene.15082 · Cited by 10
“[20] and later expanded by VanderWheele [21] has improved the understanding between disease causation and interaction, in that presence of interaction between two causal factors on the additive scale implies that there exists a pathway towards disease where the presence of both risk factors is needed (sufficient- caus…”
BackgroundA competing risks approach to “biologic” interaction
Andersen, P. K., & Skrondal, A. (2015) · Lifetime Data Analysis · doi:10.1007/s10985-015-9318-z · Cited by 9
Choi, S., Ko, K., Park, J. B., Lee, K. J., Lee, S., & Jeong, I. (2020) · Annals of Occupational and Environmental Medicine · doi:10.35371/aoem.2020.32.e33 · Cited by 10
Nova, A., Gentilini, D., Caprio, G. D., Bourguiba, S., Vince, N., Gourraud, P., Bernardinelli, L., & Fazia, T. (2025) · European Journal of Neurology · doi:10.1111/ene.70131 · Cited by 10
Gammelmark, A., Lundbye‐Christensen, S., Tjønneland, A. M., Schmidt, E. B., Overvad, K., & Nielsen, M. S. (2017) · Journal of Nutrition · doi:10.3945/jn.117.247569 · Cited by 8
Delaney, K. P., Rurangirwa, J., Facente, S. N., Dowling, T., Janson, M., Knoble, T., Vu, A., Hu, Y. W., Kerndt, P. R., King, J. B., et al. (2015) · JAIDS Journal of Acquired Immune Deficiency Syndromes · doi:10.1097/qai.0000000000000807 · Cited by 8
Combining Disease Models to Test for Gene-Environment Interaction in Nuclear Families
Hoffmann, T. J., Vansteelandt, S., Lange, C., Silverman, E. K., DeMeo, D. L., & Laird, N. M. (2011) · Biometrics · doi:10.1111/j.1541-0420.2011.01581.x · Cited by 8
“We define interaction in a statistical sense here as a departure from additive main effects in a generalized linear model (GLM), which may be different than a biological interaction (Cordell, 2002; VanderWeele, 2009).”
BackgroundRayman, J. B., Melas, P. A., Schalling, M., Forsell, Y., Kandel, E. R., & Lavebratt, C. (2019) · Journal of Affective Disorders · doi:10.1016/j.jad.2019.09.018 · Cited by 9
VanderWeele, T. J. (2012) · American Journal of Epidemiology · doi:10.1093/aje/kws214 · Cited by 8
Self-citationHe, Y., Liu, W., Lin, S., Li, Z., Qiu, H., Yim, S. H. L., Chuang, H., & Ho, K. (2023) · The Science of The Total Environment · doi:10.1016/j.scitotenv.2023.166531 · Cited by 7
Mooney, S. J., Grady, S. T., Sotoodehnia, N., Lemaître, R. N., Wallace, E. R., Mohanty, A. F., Yee, J., Siscovick, D. S., Rea, T. D., McKnight, B., et al. (2016) · Epidemiology · doi:10.1097/ede.0000000000000503 · Cited by 9
Dai, J. Y., Tapsoba, J. D. D., Buas, M. F., Risch, H. A., Vaughan, T. L., Chow, W., Shaheen, N. J., Anderson, L. A., Corley, D. A., Gammon, M. D., et al. (2016) · The American Journal of Human Genetics · doi:10.1016/j.ajhg.2016.06.018 · Cited by 7
“Much discussion has been devoted to methodological issues,(6,7) such as design and estimation, power and sample size,(12) and mathematical formulation of interactions.(13) When multiple environmental factors have been established for a cancer and are available for G 3 E testing, the standard analytical approach is to…”
BackgroundIn defense of “biological interaction”
Rajaleid, K., Janszky, I., & Hallqvist, J. (2011) · Epidemiology · doi:10.1097/ede.0b013e31820932ac · Cited by 6
“techniques used to organize the empirical data according to the needs of the criteria.(7,8) Given that the definition of”
MethodLuo, W., Shi, Z., Kong, L., Wang, X., & Zhou, H. (2023) · European Journal of Neurology · doi:10.1111/ene.16178 · Cited by 6
“We examined interaction on the additive scale based on an epidemiological definition [16].”
MethodDeep determinism and the assessment of mechanistic interaction
Berzuini, C., & Dawid, A. P. (2012) · Biostatistics · doi:10.1093/biostatistics/kxs049 · Cited by 6
Björk, J., Andersson, T., & Ahlbom, A. (2019) · European Journal of Epidemiology · doi:10.1007/s10654-019-00573-8 · Cited by 6
“1 3 disease load we have to model, or more preferably observe, their joint effects on disease [17].”
BackgroundLin, J., Lin, K.-H., Lee, W., & Lin, S. (2019) · Statistics in Medicine · doi:10.1002/sim.8401 · Cited by 6
Strength in causality: discerning causal mechanisms in the sufficient cause model
Suzuki, E., & Yamamoto, E. (2021) · European Journal of Epidemiology · doi:10.1007/s10654-021-00798-6 · Cited by 6
“Indeed, the sufficient cause model is essential to addressing important causal concepts in epidemiology, including “biologic” interaction [17, 41, 42], mediation [43–45], and etiologic fraction [35, 36, 46], which would be useful to consider the plausibility of association—the sixth of the nine Bradford Hill viewpoint…”
BackgroundThe Sufficient Cause Framework in Statistics, Philosophy and the Biomedical and Social Sciences
VanderWeele, T. J. (2012) · Wiley series in probability and statistics · doi:10.1002/9781119945710.ch13 · Cited by 6
Self-citationEstimation of a Common Effect Parameter from Follow-Up Data When There Is No Mechanistic Interaction
Lee, W. (2014) · PLoS ONE · doi:10.1371/journal.pone.0086374 · Cited by 5
“The sufficient component cause model [1–13] can help to resolve this conflict.”
BackgroundBOOLEAN ALGEBRAS AND CLASSIFICATION OF INTERACTIONS IN SUFFICIENT-COMPONENT CAUSE MODEL
Panov, V. G., & Nagrebetskaya, J. (2015) · International Journal of Pure and Apllied Mathematics · doi:10.12732/ijpam.v98i2.7 · Cited by 6
“In compliance with [10], [20], [31], [33] synergism or antagonism is present in the experiment as soon as an individual is presented with a response of types CM or CS .”
BackgroundMethodInfluentialA General Framework for and New Normalization of Attributable Proportion
Hössjer, O., Kockum, I. S., Alfredsson, L. S., Hedström, A. K., Olsson, T. P., & Lekman, M. (2016) · Epidemiologic Methods · doi:10.1515/em-2015-0028 · Cited by 5
“For binary factors and outcomes, it has been shown in VanderWeele (2009) and references therein, that under mild regularity conditions, sufficient cause interaction is closely related to additive interaction.”
BackgroundPearce, N. E., & Greenland, S. (2022) · European Journal of Epidemiology · doi:10.1007/s10654-022-00931-z · Cited by 5
“This “synergism” tells us little about the underlying biologic mechanisms, because many different biological models could produce the same risk patterns[10, 13, 18, 20].”
BackgroundHow Much is Minnesota Like Wisconsin? States as CounterfactualsPreprint
Keele, L. J., & Minozzi, W. (2010) · SSRN Electronic Journal · Cited by 5
Additive Interaction in the Presence of a Mismeasured Outcome
Jiang, Z., & VanderWeele, T. J. (2014) · American Journal of Epidemiology · doi:10.1093/aje/kwu351 · Cited by 4
Self-citationSansom, K., Mittinty, M., Dunbar, C., Vakulin, A., Adams, R. J., McArdle, N., Eastwood, P. R., & Reynolds, A. C. (2025) · Sleep Medicine · doi:10.1016/j.sleep.2025.106699 · Cited by 4
A Bayesian semiparametric approach to learning about gene–gene interactions in case-control studies
Bhattacharya, D., & Bhattacharya, S. (2018) · Journal of Applied Statistics · doi:10.1080/02664763.2018.1444741 · Cited by 4
“Although VanderWeele [29] derived some strong empirical conditions under which statistical interactions correspond to compositional epistasis, a prevailing opinion reflected in both genetic and epidemiological literature suggests limitations of the tests based on Fisher’s statistical definition of epistasis in explain…”
BackgroundChen, X., Wang, K., & Chen, D. (2020) · ICSA book series in statistics · doi:10.1007/978-3-030-35260-8_15 · Cited by 3
Pearce, N. E., & Greenland, S. (2024) · Springer eBooks · doi:10.1007/978-1-4614-6625-3_10-1 · Cited by 4
Analysis of Interaction for Identifying Causal Mechanisms
Berzuini, C., Dawid, A. P., Zhang, H., & Parkes, M. (2012) · Wiley series in probability and statistics · doi:10.1002/9781119945710.ch14 · Cited by 3
Stochastic mechanistic interaction
Berzuini, C., & Dawid, A. P. (2016) · Biometrika · doi:10.1093/biomet/asv072 · Cited by 4
“Various authors have proposed tests for inferring mechanistic interaction (suitably defined) from observational data (Rothman, 1976; Rothman & Greenland, 1998; Greenland & Poole, 1988; VanderWeele & Robins, 2008, 2009; VanderWeele, 2009,2 10, 2011; VanderWeele & Laird, 2011).”
BackgroundValidity and bias in epidemiological research
Greenland, S., & VanderWeele, T. J. (2015) · doi:10.1093/med/9780199661756.003.0116 · Cited by 3
Self-citationFaquetti, M. L., Vallejo‐Yagüe, E., Cordtz, R., Dreyer, L., & Burden, A. M. (2023) · PLoS ONE · doi:10.1371/journal.pone.0288757 · Cited by 2
“Within the corresponding exposure set, each JAKi user will be matched to one TNF-αi user at T0 using time-conditional propensity scores (TCPS).”
MethodInfluentialNova, A., Fazia, T., & Bernardinelli, L. (2024) · Medicina · doi:10.3390/medicina60071053 · Cited by 4
“As illustrated in [2,14,15], assessing additive interaction can be used to obtain insights on the presence of interaction in a mechanistic sense other than from a purely statistical point of view.”
BackgroundHildt, C., Tzivian, L., Ogurtsova, K., & Hoffmann, B. H. (2025) · Environmental Research · doi:10.1016/j.envres.2025.123589 · Cited by 2
Prevalence of Congenital Heart Defects in Neonates of Drug Abusing Mothers
Radgoodarzi, M., Norouzi, E., Vahedi, Z., Salavati, M., Yaghoubi, A., & Behmadi, R. (2021) · international journal high risk behaviors & addiction · doi:10.5812/ijhrba.106884 · Cited by 2
Jääskeläinen, T., Koponen, P., Lundqvist, A., Suvisaari, J., Järvelin, J., & Koskinen, S. V. P. (2022) · BMJ Open · doi:10.1136/bmjopen-2021-056073 · Cited by 2
Sim, K., & Eastern, D.-J. (2016) · Cited by 2
Bayesian Exploration of Multilocus Interactions on the Genome-Wide Scale
Mohammad, M. (2012) · Current Research in Bioinformatics · doi:10.3844/ajbsp.2012.70.78 · Cited by 2
“It has been even suggested that functional epistasis might not be detectable in the current GWAS as statistical interactions (Greenland, 2009; Vanderweele, 2009).”
BackgroundCompositional Epistasis: An Epidemiologic Perspective
Suzuki, E., & VanderWeele, T. J. (2014) · Methods in molecular biology · doi:10.1007/978-1-4939-2155-3_11 · Cited by 3
Self-citationNSAIDs, hormonal contraception, and venous thromboembolism
Schmidt, M. (2023) · BMJ · doi:10.1136/bmj.p1990 · Cited by 2
A Framework for Thinking About the Potential Public Health Impact of Epidemiologic Research
Lesko, C. R., Zalla, L. C., Ross, R. K., Rudolph, J. E., Smith, E. R., & Edwards, J. K. (2025) · Epidemiology · doi:10.1097/ede.0000000000001939 · Cited by 1
Mena, S. P. Q., Piñeros, J. G., & Cañón-Montañéz, W. (2024) · Tropical Medicine and Infectious Disease · doi:10.3390/tropicalmed9090196 · Cited by 1
“Renowned authors such as Rothman [35] and VanderWeele [36] have made contributions to direct the analysis of this element, providing foundations for its theoretical distinction and the implementation of formal tests to analyze the interaction within statistical models, ranging from interaction in…”
BackgroundJiang, X. (2015) · Figshare · Cited by 2
Atashili, J., & Kaufman, J. S. (2013) · American Journal of Epidemiology · doi:10.1093/aje/kwt202 · Cited by 2
Hosmer, D. W., Lemeshow, S. A., & Sturdivant, R. X. (2013) · Wiley series in probability and statistics · doi:10.1002/9781118548387.refs · Cited by 1
McGrane, M., & Poon, S. K.-S. (2011) · AIS Electronic Library (AISeL) · Cited by 3
van de Watering, L. M. G., & Middelburg, R. A. (2010) · Epidemiology · doi:10.1097/ede.0b013e3181e942a2
Genes, lifestyle and coronary heart disease risk-epidemiological interaction studies
Gustavsson, J. (2015) · Gothenburg University Publications Electronic Archive (Gothenburg University)
Sharp bounds on sufficient-cause interactions under the assumption of no redundancy
Lee, W. (2017) · BMC Medical Research Methodology · doi:10.1186/s12874-017-0348-y
Testing for Sufficient-Cause Interactions in Case-Control Studies of Non-Rare Diseases
Lin, J., & Lee, W. (2018) · Scientific Reports · doi:10.1038/s41598-018-27660-2
Unraveling gene gene interactions in rheumatoid arthritis
Lodhi, S. (2021) · University Library of Skövde (University of Skövde)
Validity and bias in epidemiological research
Greenland, S., & VanderWeele, T. J. (2021) · doi:10.1093/med/9780198816805.003.0036
Self-citationHsia, P.-H., Tai, A., Fu, S., & Lin, S. (2025) · Statistical Methods in Medical Research · doi:10.1177/09622802251376236
“24 In the additive scale, let T SCI = p 11 − p 10 − p 01 , where p x 1 x 2 ≡ Pr ( Y = 1 X 1 = x 1 , X 2 = x 2 ) and the test statistic ̂ T SCI = ̂ p 11 − ̂ p 10 − ̂ p 01 where p x 1 x 2 can be estimated by the empirical mean of Y of a stratum ( X 1 , X 2 ) = ( x 1 , x 2 ) .”
Hou, H., Zha, X., Lu, Z., Chen, P., Wang, L., Wang, F., & Wang, R. (2025) · BMC Pregnancy and Childbirth · doi:10.1186/s12884-025-08483-2
“If the 95% confidence interval (CI) of RERI excluded 0, additive interaction was present; if the P-value of the product term was < 0.05, multiplicative interaction was indicated [23–25].”
InfluentialCanoy, D., Tran, J., Zottoli, M., Ramakrishnan, R., Hassaïne, A., Rao, S., Li, Y., Salimi‐Khorshidi, G., Norton, R., & Rahimi, K. (2021) · Figshare · doi:10.6084/m9.figshare.16892983
Lalloué, B. (2013) · theses.fr (ABES) · doi:10.70675/bed1b010z8b1az43e6z87b2z26c5e0605631 · Cited by 1
Khan, P. Y., Govender, I., McCreesh, N., Sithole, M., Mkwanzai, E., Sweeney, S., Ording-Jespersen, G., Wong, E. B., Hanekom, W. A., Houben, R. M. G. J., et al. (2026) · medRxiv · doi:10.64898/2026.06.01.26354571 · Cited by 1
Metadata
- Authors
- VanderWeele, T. J.
- Affiliations
- University of Chicago (US)
- Corresponding author
- VanderWeele, Tyler J.
- Venue
- Epidemiology
- Published
- 2009-01
- Volume
- 20
- Issue
- 1
- Pages
- 6-13
- ISSN
- 1044-3983 (ISSN-L)
- OpenAlex
- W2056949311
- Cited by
- 163
- References
- 20
- Citations by year
- 2026: 2 · 2025: 5 · 2024: 5 · 2023: 8 · 2022: 4 · 2021: 12 · 2020: 8 · 2019: 5 · 2018: 5 · 2017: 10 · 2016: 6 · 2015: 13 · 2014: 9 · 2013: 10 · 2012: 22
- FWCI
- 12.28
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Epidemiology (publishedVersion)
- PubMed (publishedVersion)
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Epidemiology
- Physical Sciences › Computer Science › Artificial Intelligence › Bayesian Modeling and Causal Inference
- Keywords
- statistical interaction, binary outcomes, interaction parameters, logistic regression, log-linear models, confounding
- MeSH terms
- Algorithms, Data Interpretation, Statistical, Causality, Epidemiologic Studies
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.