Interactions in Epidemiology: Relevance, Identification, and Estimation
The DOI matches its Crossref record; 29 of the 30 references it cites were found in a registry or an index.
Abstract
Sufficient-cause models represent biologic interactions (including synergism) in the form of coparticipation of factors in a sufficient cause1–3 (causal coaction). In this issue of the journal, VanderWeele1 provides new results relating sufficient-cause models to regression models. These results reinforce the point that concepts of biologic interaction do not in general correspond to the concept of statistical interaction, because the latter is only the need for a product term in a statistical model.4,5 They further underscore that, without strict assumptions, sufficient-cause interactions do not correspond to interdependence of effects in potential-outcomes (counterfactual) models. I will examine the practical relevance and limitations of using epidemiologic regressions to study interactions, however the latter are defined. Relevance of Interactions A practical issue concerning interactions is their relevance to public health goals of finding cost-effective intervention strategies for disease reduction. Part of public health involves identifying population subgroups that would benefit most from a given intervention. These are not always high-risk subgroups—it is possible that a lower-risk group would obtain the greatest absolute risk reduction from the proposed intervention. In the absence of bias, departures from risk additivity imply that some subgroups would obtain a greater absolute risk reduction from the intervention than others would. Thus earlier authors noted that if costs were measured in terms of case-load per unit population (average risk), the relevant null model for detecting special groups would be additive on the risk scale.6,7 Risk nonadditivity (or “public health interaction”) implies the existence of such groups.5,6 It also implies existence of interdependent effects in individuals, also known as “interactions” in basic potential-outcome (counterfactual) causal models; that is, risk nonadditivity implies there are individuals such that the presence or direction of effect of one factor depends on the other factor.5,8 Risk nonadditivity is equivalent to heterogeneity or modification of risk differences for one variable across levels of another.5 For factors that can act only causally, excess risk above additivity (superadditivity) among those exposed to both factors signals the presence of individuals who get the disease only when exposed to both factors (“synergism” in the potential-outcomes sense).4,5,8 Identifying such individuals is valuable for preventive medicine and public health, because for them the outcome can be prevented by removing either of the factors. Nonidentifiability of Biologic Interactions Epidemiologic studies observe population distributions of risks, which submerge details of individual experiences. One consequence of this submersion is that many different biologic models will lead to the same population model.9,10 In particular, the implications of biologic models for interaction are unidirectional; for example, risk additivity could hold because no interdependence of effects exists, or instead may hold because of cancellations of the population effects of different individual interactions.5,8 Perfect cancellation of interactions is a higher-order analog of “unfaithfulness” in causal diagrams,11,12 where 2 variables may be unassociated despite having causal connections. For example, a treatment with effects could appear independent of the outcome if it causes and prevents equal numbers of cases. Although perfect cancellation is often implausible, any degree of cancellation will make it more difficult to detect effects or their interdependencies. Even if we could observe each individual's response under each exposure pattern, we still could not determine from those observations which biologic mechanisms or sufficient-cause types were operative8,13–15; hence, individual mechanisms are nonidentifiable even from a perfect factorial or crossover trial.13,15 For example, even with no sufficient-cause synergism, variation in background characteristics could nonetheless produce superadditivity.14 More generally, estimation of degrees of interaction will require further assumptions (eg, monotonicity) that cannot be tested with epidemiologic data, even if the latter are perfect, although valid tests for interactions can be constructed from weaker conditions.1,5,8,13–15 Statistical Modeling Issues Epidemiologic theories of interactions1,4,5,8,9,14–17 derive population causal models from models for individual etiology (eg, risk additivity is derived from the absence of interdependent effects). Study of these models and forms does not require that one use the derived population model as the statistical analysis model. One can estimate risks or rates by using any statistical model, then use these estimates to test or estimate the parameters in the causal model.1,18 For example, one can use risk estimates from a logistic statistical model for inference on parameters in an additiv
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
“Regarding studies that employ the product term specification, epidemiologists have called for an end to the practice of assessing interactions based purely on product terms in multiplicative models (Greenland, 2009; Rajaleid, Janszky, & Hallqvist, 2011).”
From Syndemics of psychosocial problems and HIV risk: A systematic review of empirical tests of the disease interaction concept (2015) · cites this paper as Greenland, 2009
In the citing paper
“The concept of disease interaction has had a long and controversial history in the epidemiologic literature (Blot & Day, 1979; Greenland, 2009; Kaufman, 2009; Kupper & Hogan, 1978; Rothman, 1974, 1976a; Rothman, Greenland, & Walker, 1980; Saracci, 1980; Siemiatycki & Thomas, 1981; VanderWeele, 2009a; Walter & Holford, 1978).”
From Syndemics of psychosocial problems and HIV risk: A systematic review of empirical tests of the disease interaction concept (2015) · cites this paper as Greenland, 2009
In the citing paper
“At the same time, from the public health perspective, additive effects are favourable over multiplicative effects to increase the net benefit by allocating the treatment to the proper subgroup.12,13”
From Understanding of interaction (subgroup) analysis in clinical trials (2019) · cites this paper as [12]
Citations
Cites
The Elements of Statistical Learning: Data Mining, Inference, and Prediction
John Lu, Z. Q. (2010) · Journal of the Royal Statistical Society Series A: Statistics in Society · doi:10.1111/j.1467-985x.2010.00646_6.x · Cited by 12,390
Bootstrap Methods and their Application
Davison, A. C., & Hinkley, D. V. (1997) · doi:10.1017/cbo9780511802843 · Cited by 7,350
Friedman, J., Hastie, T., & Tibshirani, R. (2000) · The Annals of Statistics · doi:10.1214/aos/1016218223 · Cited by 6,884 · cited 1×
“included, this goal leads to fitting methods that can use a very large set of terms, such as shrinkage(27) and machine-learning algorithms such as boosting.(28,29) The same methods can also”
MethodROTHMAN, K. J., GREENLAND, S., & WALKER, A. M. (1980) · American Journal of Epidemiology · doi:10.1093/oxfordjournals.aje.a113015 · Cited by 859 · cited 2×
“Thus earlier authors noted that if costs were measured in terms of case-load per unit population (average risk), the relevant null model for detecting special groups would be additive on the risk scale.(6,7) Risk nonadditivity (or “public health interaction”) implies the existence of such groups.”
BackgroundSelf-citationROTHMAN, K. J. (1976) · American Journal of Epidemiology · doi:10.1093/oxfordjournals.aje.a112335 · Cited by 776
Multiple-bias modelling for analysis of observational data
Greenland, S. (2005) · Journal of the Royal Statistical Society Series A (Statistics in Society) · doi:10.1111/j.1467-985x.2004.00349.x · Cited by 497
Self-citationTHE EFFECT OF MISCLASSIFICATION IN THE PRESENCE OF COVARIATES
GREENLAND, S. (1980) · American Journal of Epidemiology · doi:10.1093/oxfordjournals.aje.a113025 · Cited by 494 · cited 1×
“ity (such as nonadditivity), and measurement error can cause any of these problems even if it is nondifferential.(21,32)”
BackgroundSelf-citationAn 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 · cited 1×
“Even if we could observe each individual’s response under each exposure pattern, we still could not determine from those observations which biologic mechanisms or sufficient-cause types were operative; hence, individual mechanisms are nonidentifiable even from a perfect factorial or crossover trial.(13,15) For example…”
BackgroundSelf-citationInvited Commentary: Variable Selection versus Shrinkage in the Control of Multiple Confounders
Greenland, S. (2007) · American Journal of Epidemiology · doi:10.1093/aje/kwm355 · Cited by 283
Self-citationTests for interaction in epidemiologic studies: A review and a study of power
Greenland, S. (1983) · Statistics in Medicine · doi:10.1002/sim.4780020219 · Cited by 280
Self-citation
Research works printed but not linked: 1
- Greenland, 2007, Encyclopedia of Epidemiology, Vol. 1., 149
Cited by
Items 101–200 of 203
The Importance of Effect Measure Modification When Using Demographic Variables to Predict Evacuation
Horney, J. A., MacDonald, P. D. M., Willigen, M. V., & Kaufman, J. S. (2012) · Risk Hazards & Crisis in Public Policy · doi:10.1515/1944-4079.1084 · Cited by 13
Pearce, N. E., & Greenland, S. (2013) · Springer eBooks · doi:10.1007/978-0-387-09834-0_10 · Cited by 14
Self-citationLi, M., Lou, X.-Y., & Lu, Q. (2012) · Recent Patents on Biotechnology · doi:10.2174/1872208311206030230 · Cited by 15
“In epidemiology, the interaction arises when disease risks that are attributable to two or more risk factors are greater than the additive or multiplicative of each factor alone [10].”
BackgroundMadsen, I. E. H., Larsen, A. D., Thorsen, S. V., Pejtersen, J. H., Rugulies, R., & Sivertsen, B. (2016) · Scandinavian Journal of Work Environment & Health · doi:10.5271/sjweh.3571 · Cited by 14
Madsen, I. E. H., Jørgensen, A. F. B., Borritz, M., Nielsen, M. L., & Rugulies, R. (2014) · BMC Public Health · doi:10.1186/1471-2458-14-698 · Cited by 13
“We base our conclusions regarding effect-modification on departure from additivity because such departure is particularly relevant from the clinical and public health perspectives and identifies groups who benefit most from exposure intervention [16,38-40].”
BackgroundUng, L., VanderWeele, T. J., & Dahabreh, I. J. (2025) · Epidemiology · doi:10.1097/ede.0000000000001863 · Cited by 16
“As usual, interpreting the results of empirical examinations of interaction requires considering the potential for limited precision, the scale on which interaction is assessed, and model specification [49–52].”
BackgroundXiu-ling, Z., Cao, J., Zhou, W., & Neophytou, A. M. (2025) · Journal of the American Heart Association · doi:10.1161/jaha.124.037329 · Cited by 11
Semiparametric Tests for Sufficient Cause Interaction
Vansteelandt, S., VanderWeele, T. J., & Robins, J. M. (2012) · Journal of the Royal Statistical Society Series B (Statistical Methodology) · doi:10.1111/j.1467-9868.2011.01011.x · Cited by 11
“E-mail: Stijn.Vansteelandt@ugent.be VanderWeele, 2009; Greenland, 2009).”
Conditions for confounding of interactions
Liu, A., Abrahamowicz, M., & Siemiatycki, J. (2015) · Pharmacoepidemiology and Drug Safety · doi:10.1002/pds.3924 · Cited by 11
Chen, M., Shi, W., Wang, H., Zhao, L., Guo, X., & Sun, Y. (2021) · Diabetes Metabolic Syndrome and Obesity · doi:10.2147/dmso.s295960 · Cited by 10
Pollet, T. V., Dawson, S., Tovée, M. J., Cornelissen, P. L., & Cornelissen, K. K. (2021) · Body Image · doi:10.1016/j.bodyim.2021.05.005 · Cited by 11
McRae, D. N., Muhajarine, N., & Janssen, P. A. (2019) · BMC Pregnancy and Childbirth · doi:10.1186/s12884-019-2428-y · Cited by 13
Nkwata, A. K., Zhang, M., Song, X., Giordani, B., & Ezeamama, A. E. (2022) · International Journal of Environmental Research and Public Health · doi:10.3390/ijerph19042419 · Cited by 10
“10 because the power of statistical tests for higher-order terms is generally lower than for first-order terms [34,35].”
BackgroundWhitcomb, B. W., & Naimi, A. I. (2023) · American Journal of Epidemiology · doi:10.1093/aje/kwad034 · Cited by 10
Probst, C., Zhu, Y., Kilian, C., Kerr, W. C., & Rehm, J. T. (2025) · Addiction · doi:10.1111/add.16774 · Cited by 9
“Departure from additivity of effects implies that certain population subgroups would gain a larger absolute risk reduction from the intervention than others [14].”
Timmermans, E. J., Veldhuizen, E. M., Mäki-Opas, T. E., Snijder, M. B., Lakerveld, J., & Kunst, A. E. (2019) · Spatial and Spatio-temporal Epidemiology · doi:10.1016/j.sste.2019.100300 · Cited by 10
Neonatal Jaundice and Autism: Precautionary Principle Invocation Overdue
Wilde, V. K. (2022) · Cureus · doi:10.7759/cureus.22512 · Cited by 10
“identification of specific mechanisms of interaction will depend on biologic assumptions that are untestable (nonidentifiable) with epidemiologic observations alone" [252].”
BackgroundBloome, D., & Ang, S. (2022) · Demography · doi:10.1215/00703370-10109444 · Cited by 11
Shusterman, M., Golub, E., Mowrey, W. B., & Broder, A. R. (2017) · Lupus · doi:10.1177/0961203317721352 · Cited by 10
Candidate gene–environment interactions: Table 1
Kauffmann, F., & Nadif, R. (2010) · Journal of Epidemiology & Community Health · doi:10.1136/jech.2008.086199 · Cited by 9
“The use of the term interaction often confuses statistical interaction (product term in a model) with biological interaction and it has even been proposed to avoid that term in gene environment interactions (12).”
BackgroundKondracki, A. J., Attia, J. R., Valente, M. J., Roth, K. B., Akin, M., McCarthy, C., & Barkin, J. L. (2024) · Neuropsychiatric Disease and Treatment · doi:10.2147/ndt.s450236 · Cited by 8
“24,26,32 The structure of a DAG provided a graphical representation of the “backdoor” criterion 24 with several paths by which the effect of exposure to smoking during pregnancy and the presence of antenatal depression may be causally related to PPD (Figure 2).”
BackgroundTejera, C. H., Ware, E. B., Kobayashi, L. C., Fu, M., Hicken, M. T., Zawistowski, M., Mukherjee, B., & Bakulski, K. M. (2023) · Frontiers in Human Neuroscience · doi:10.3389/fnhum.2023.1052435 · Cited by 8
“Nonetheless, epidemiologists prefer estimating interaction effects in the additive scale (Greenland, 2009).”
BackgroundVanderWeele, T. J. (2012) · American Journal of Epidemiology · doi:10.1093/aje/kws214 · Cited by 8
The role of linkage disequilibrium in case-only studies of gene–environment interactions
Yadav, P., Freitag‐Wolf, S., Lieb, W., & Krawczak, M. (2014) · Human Genetics · doi:10.1007/s00439-014-1497-2 · Cited by 7
Socioeconomic position, alcohol use and alcohol-attributable emergency department visits
Smith, B. T., Benny, C., Andreacchi, A. T., Schwartz, N., Warren, C. M., Forbes, S. M., & Hobin, E. P. (2024) · Journal of Epidemiology & Community Health · doi:10.1136/jech-2024-222476 · Cited by 6
Holakouie‐Naieni, K., Nematollahi, S., Mansournia, M. A., Shekari, M., Agha-Molayi, T., Alavi, A., Madani, A.-H., Holakouie-Naieni, N., Shabkhiz, H., Torabi, E., et al. (2018) · PubMed Central · Cited by 9
The association of milk products with rheumatoid arthritis: A cross-sectional study from NHANES
Chen, W., Jiang, D., Liu, K., Lyu, L., Chen, Y., Sun, X., Mao, Y., & Ye, D. (2023) · Joint Bone Spine · doi:10.1016/j.jbspin.2023.105646 · Cited by 6
Lefebvre, G., & Gustafson, P. (2010) · The International Journal of Biostatistics · doi:10.2202/1557-4679.1207 · Cited by 6
Case-only analysis of gene–gene interactions in inflammatory bowel disease
Aleknonytė‐Resch, M., Freitag‐Wolf, S., The International Inflammatory Bowel Disease Genetics Consortium, Schreiber, S., Krawczak, M., & Dempfle, A. (2020) · Scandinavian Journal of Gastroenterology · doi:10.1080/00365521.2020.1790646 · Cited by 6
Hurricane evacuation failure: the role of social cohesion, social capital, and social control
Horney, J. A. (2009) · Carolina Digital Repository (University of North Carolina at Chapel Hill) · doi:10.17615/x9cj-w198 · Cited by 7
Genetic and early life factors influence on time-to-multiple sclerosis diagnosis: A UK Biobank study
Nova, A., Caprio, G. D., Bernardinelli, L., & Fazia, T. (2024) · Multiple Sclerosis Journal · doi:10.1177/13524585241257205 · Cited by 5
“We then predicted the MS diagnosis CINC, by sex, based on different combinations of MS-PRS, smoking status, and previous IM diagnosis.”
InfluentialWang, S., Chen, W. J., Chuang, L., Hsiao, P., Liu, P., & Hsiao, C. K. (2013) · PLoS ONE · doi:10.1371/journal.pone.0056693 · Cited by 6
Bally, M., Nadeau, L., & Brophy, J. M. (2018) · PLoS ONE · doi:10.1371/journal.pone.0201884 · Cited by 5
Responding to Repeated Disasters: Time to Recovery in Public Health Workers
Mash, H. B. H., Fullerton, C. S., Morganstein, J. C., Liu, A. G., Vance, M. C., Wang, L., Mullins-Hussain, B., & Ursano, R. J. (2022) · Disaster Medicine and Public Health Preparedness · doi:10.1017/dmp.2022.130 · Cited by 6
Nova, A., Fazia, T., & Bernardinelli, L. (2024) · Medicina · doi:10.3390/medicina60071053 · Cited by 4
“While guidelines suggest reporting interaction using both scales [2,3], evaluating interactions on an additive scale is considered more relevant in the context of disease prevention, as it is more suitable to determine how best to allocate medical interventions and treatments [4].”
BackgroundRelevance and challenges of exposome studies for environmental health research
Slama, R., Siroux, V., Vrijheid, M., & Basagaña, X. (2026) · Environmental Health · doi:10.1186/s12940-026-01299-3 · Cited by 5
“In addition, toxicologists and epidemiologists tend to have different views on synergy, and, as already discussed, epidemiologists have long conflated synergy with statistical interactions [5, 157].”
Madsen, I. E. H., Hannerz, H., Nyberg, S. T., Hanson, L. L. M., Ahola, K., Alfredsson, L. S., Batty, G. D., Bjørner, J. B., Borritz, M., Burr, H., et al. (2013) · F1000Research · doi:10.12688/f1000research.2-233.v2 · Cited by 5
Gefährdungsbeurteilung bei Exposition gegenüber mehreren krebserzeugenden Arbeitsstoffen
Morfeld, P., & Spallek, M. (2015) · Zentralblatt für Arbeitsmedizin Arbeitsschutz und Ergonomie · doi:10.1007/s40664-015-0032-1 · Cited by 4
Neophytou, A. M., Aalborg, J., Magzamen, S. L., Moore, B. F., Ferrara, A. M., Karagas, M. R., Trasande, L., & Dabelea, D. (2024) · Environmental Health Perspectives · doi:10.1289/ehp13961 · Cited by 5
Stoicescu, C., Medley, B., Wu, E., El‐Bassel, N., Tanjung, P., & Gilbert, L. (2024) · International Journal of Drug Policy · doi:10.1016/j.drugpo.2024.104486 · Cited by 5
Pearce, N. E., & Greenland, S. (2024) · Springer eBooks · doi:10.1007/978-1-4614-6625-3_10-1 · Cited by 4
Self-citationRatio of Observed and Expected Joint Effects
Gebrekristos, H. T., & Howe, C. J. (2014) · Epidemiology · doi:10.1097/ede.0000000000000200 · Cited by 4
Prasad, S., Murphy, S. A., Morrow, D. A., Scirica, B. S., Sabatine, M. S., Berg, D. D., & Bellavia, A. (2025) · Annals of Epidemiology · doi:10.1016/j.annepidem.2025.10.012 · Cited by 3
Zhu, Y., Kilian, C., Lemp, J. M., Llamosas‐Falcón, L., & Probst, C. (2025) · International Journal of Epidemiology · doi:10.1093/ije/dyaf190 · Cited by 3
Statistical Aspects of Racial and Ethnic Health Disparities
Kaufman, J. S. (2025) · Annual Review of Statistics and Its Application · doi:10.1146/annurev-statistics-042324-061403 · Cited by 5
The geometry of G × E: How scaling and endogenous treatment effects shape interaction direction
Sadowski, M., Dahl, A. W., Zaitlen, N., & Border, R. (2026) · PLoS Genetics · doi:10.1371/journal.pgen.1012073 · Cited by 3
“In case (1), the detection of an interaction effect and its direction depend on the scale on which the outcome is measured, or to which it might be transformed [18–21].”
Effect of Arsenic Exposure and Cigarette Smoking on Total and Cause-Specific Mortality
Jia, X., Li, Y.-F., Fan, Y.-G., Zhou, Q.-H., Zhao, F.-H., Qiao, Y.-L., & Lalley, M. (2023) · Journal of Occupational and Environmental Medicine · doi:10.1097/jom.0000000000002764 · Cited by 2
Nkwata, A. K., & Smith, J. (2023) · Health Science Reports · doi:10.1002/hsr2.1756 · Cited by 3
“Since these moderation analyses were exploratory and expected to be underpowered, interaction terms with p ‐ values 0.10 were considered potentially important, 49 and stratum specific results presented where potential interactions were indicated.”
BackgroundInfluentialde Campos, M. C. T., Rosa, R. J., Ferezin, L. P., Berra, T. Z., Moura, H. S. D., Tártaro, A. F., Alves, Y. M., Tavares, R. B. V., Ribeiro, N. M., de Araújo, J. S. T., et al. (2025) · BMC Women s Health · doi:10.1186/s12905-025-03793-1 · Cited by 4
“This is a cross-sectional and quantitative study [26], in which data was collected using a questionnaire adapted and validated for Brazil.”
Carrillo-Balam, G., Li, Y., & Silverman-Retana, O. (2023) · Journal of Epidemiology & Community Health · doi:10.1136/jech-2022-219557 · Cited by 3
Bonifaci, N., Colás, E., Serra-Musach, J., Karbalai, N., Brunet, J., Gómez, A., Esteller, M., Fernández-Taboada, E., Berenguer, A., Reventós, J., et al. (2013) · Carcinogenesis · doi:10.1093/carcin/bgt403 · Cited by 3
Nkwata, A. K., Song, X., Zhang, M., & Ezeamama, A. E. (2020) · BMC Public Health · doi:10.1186/s12889-020-09399-6 · Cited by 3
Campbell, D. A., Boyle, L., Webb, M., Mistry, M., & Short, T. G. (2024) · Anesthesiology · doi:10.1097/aln.0000000000005233 · Cited by 2
Ziyab, A. H., Almari, M., Mohammad, A., Al‐Taiar, A., & Karmaus, W. (2022) · International Journal of Endocrinology · doi:10.1155/2022/8727922 · Cited by 2
“Given that statistical power to detect higher-order terms is usually limited in epidemiologic studies [34, 35], the interaction term P-value (Pinteraction)< 0.”
BackgroundDiabetes and the Hispanic Health Paradox: insights from Hispanics in Canada
Escobar, K. M., & Gorey, K. M. (2025) · Ethnicity and Health · doi:10.1080/13557858.2025.2550679 · Cited by 2
“Since this study conducted secondary analyses of an available sample, statistical power calculations were completed post hoc (Cohen 1988; Faul et al. 2007; 2009; 2013).”
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).”
BackgroundIzumi, S., Sakata, R., Yamada, M., & COLOGNE, J. B. (2015) · Journal of Clinical Epidemiology · doi:10.1016/j.jclinepi.2015.10.004 · Cited by 3
Fujii, Y., Henmi, M., & Fujita, T. (2012) · Statistics in Medicine · doi:10.1002/sim.4400 · Cited by 3
Late-Life Depressive Symptoms, Religiousness, and Mood in the Last Week of Life
Braam, A. W., Klinkenberg, M., Galenkamp, H., & Deeg, D. J. H. (2012) · Depression Research and Treatment · doi:10.1155/2012/754031 · Cited by 2
Sim, K., & Eastern, D.-J. (2016) · Cited by 2
Atashili, J., & Kaufman, J. S. (2013) · American Journal of Epidemiology · doi:10.1093/aje/kwt202 · Cited by 2
Physical Activity : The interplay between individual and neighbourhood factors
Beenackers, M. A. (2013) · RePub (Erasmus University Rotterdam) · Cited by 2
Banack, H. R., Wactawski‐Wende, J., Ochs‐Balcom, H. M., Feliciano, E. M. C., Caan, B. J., Lee, C., Anderson, G. L., Shankaran, M., & Evans, W. J. (2024) · PLoS ONE · doi:10.1371/journal.pone.0300140 · Cited by 3
“These models will include a product interaction term (muscle mass x cancer survivorship status) to separately estimate these relationships among cancer survivors and non-cancer controls [38–40].”
Akrami, R., Hadji, M., Rashidian, H., Nazemipour, M., Naghibzadeh‐Tahami, A., Ansari‐Moghaddam, A., Zendehdel, K., & Mansournia, M. A. (2024) · Global Epidemiology · doi:10.1016/j.gloepi.2024.100182 · Cited by 1
“The presence of a super-additive interaction between these exposures suggests that an intervention targeting cigarette smokers who use opium would yield greater benefits compared to those who do not [13,17].”
Yin, L., Wang, X., & Ye, W. (2016) · Communications in Statistics - Simulation and Computation · doi:10.1080/03610918.2016.1230213 · Cited by 1
“…such as ratio of the odds ratios, but the additive measure INT is of particular interest because its expression in terms of risk differences has a close relationship with some well-known classification of biological mechanisms (Greenland, 2009; Rothman et al., 2008; VanderWeele and Robins, 2007).”
BackgroundInfluentialWilliamson, B. D., Coley, R. Y., Hsu, C. W., McCracken, C., & Cook, A. J. (2023) · Prevention Science · doi:10.1007/s11121-023-01606-1 · Cited by 1
“Statistical approaches to assessing HTE often involve modeling interactions between variables (Greenland, 2009).”
BackgroundFacebook Social Use and Anxiety: A Replication Attempt
Sillence, E., Saxton, T. K., & Pollet, T. V. (2021) · Human Communication & Technology · doi:10.17161/hct.v2i1.14847 · Cited by 1
Meyer, C. S. (2016) · University of Minnesota Digital Conservancy (University of Minnesota) · Cited by 2
Obesity's role in MS: An intervention target with possible interactions
Fitzgerald, K. C., & Mowry, E. M. (2020) · Neurology Neuroimmunology & Neuroinflammation · doi:10.1212/nxi.0000000000000932 · Cited by 1
“, obtain a greater absolute risk reduction) from a specific intervention than other individuals would.(1) In this case, the observed departure from additivity associated with obesity among EBV-infected or HLA-DRB1*15:01 carriers implies that reduction in body weight would be most beneficial for those with a previous h…”
BackgroundPrzepiorkowski, J. (2021) · Brock University Digital Repository (Brock University) · Cited by 1
ROLE OF ALLERGIC SENSITIZATION, FILAGGRIN VARIANTS, AND DNA
Ziyab, A. H. (2014) · Scholar Commons (University of South Carolina)
McGrane, M., & Poon, S. K.-S. (2011) · AIS Electronic Library (AISeL) · Cited by 3
Wang, X., Ye, W., & Yin, L. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1501.05121
“INT is also called the biological interaction because its expression in terms of risk differences has a close relationship with some well-known classification of biological mechanisms [3, 4].”
BackgroundInteraction of Genetic Susceptibility and Traffic- Related Air Pollution in Cardiovascular Disease
Levinsson, A. (2015) · Gothenburg University Publications Electronic Archive (Gothenburg University)
The Genetics of Blood Pressure and Body Mass Index: The Tale of Two Major Risk Factors.
Wu, C. (2013) · Deep Blue (University of Michigan)
Public Health Genomics: Exploiting SNP-NSAID interactions to prevent colorectal cancer
Demarsh, P. A. (2011) · Library and Archives Canada (Government of Canada) · doi:10.20381/ruor-4615
“Greenland and others [31] have argued that biological interaction between risk factors is best measured as the deviation from additivity, rather than deviation from a multiplicative model.”
BackgroundLi, Q. (2011) · University of Groningen research database (University of Groningen / Centre for Information Technology)
Bellavia, A., Bottai, M., & Orsini, N. (2022) · Epidemiology Biostatistics and Public Health · doi:10.2427/11841
“Interaction is commonly assessed as a departure from additivity or multiplicativity of the effects and it has been showed that absence of interaction on one scale is likely to imply the presence of interaction on the other scale [34].”
BackgroundWang, X., Ye, W., & Yin, L. (2016) · Journal of Applied Statistics · doi:10.1080/02664763.2016.1257587
GENETIC INTERACTIONS FROM FIRST PRINCIPLESPreprint
Fernández-de-Cossío, J., Fernández-de-Cossio-Díaz, J., & Perera, Y. (2017) · arXiv (Cornell University) · doi:10.13140/rg.2.2.16941.31209
Yin, L., Ye, W., & Wang, X. (2015) · Statistics in Medicine
Huang, J., Kockum, I. S., & Stridh, P. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2008.01479
“se. This is done for a variety of reasons; it could for example be to maximize the benefit of public health interventions by identifying at-risk groups or to better understand the etiology of disease [1]. Interactions can be investigated on different scales, usually on the additive and multiplicative scale [1]. It is…”
BackgroundA self-consistent probabilistic formulation for inference of interactions
Fernández-de-Cossio-Díaz, J., Fernández-de-Cossio-Díaz, J., & Perera, Y. (2020) · Scientific Reports · doi:10.1038/s41598-020-78496-8
Facebook: Social Uses and Anxiety - A Replication AttemptPreprint
Sillence, E., Saxton, T. K., & Pollet, T. V. (2019) · PsyArXiv (OSF Preprints) · doi:10.31234/osf.io/bvgxy
“Confounding can occur where predictors correlate, and both confounding (e.g., Greenland, 2009) and measurement error in the predictors (e.g., Jaccard & Turrisi, 2003; Muff & Keller, 2015; Whisman & McClelland, 2005) could lead to significant interaction effects in regression models.”
BackgroundGasparin, M., Scarpa, B., & Stanghellini, E. (2025) · Statistica Neerlandica · doi:10.1111/stan.12369 · Cited by 2
“This investigation involves the notion of collapsibility of the effects (Greenland, 2009).”
InfluentialMuroi, K., Morita, E., Matsumoto, S., Ishihara, A., Hasegawa, S., Ishitsuka, M., Hori, D., Doki, S., Takahashi, T., Sasahara, S., et al. (2025) · JMIR Formative Research · doi:10.2196/73784
“It is important to examine interactions in epidemiology to consider effective interventions to improve outcomes with limited resources [29].”
Lan, Y., Wu, D., Feng, B., Ding, X., Xu, L., Zheng, H., Wu, K., Li, X., Balmer, L. A., Wang, W., et al. (2025) · Diabetes Obesity and Metabolism · doi:10.1111/dom.70274
“The presence of supra-additive interactions is of great significance from a public health standpoint, 45 informing health strategies against the type 2 diabetes epidemic.”
Roa‐Díaz, Z. M., Teuscher, J., Gamba, M., Bundo, M., Grisotto, G., Wehrli, F., Gamboa, E., Rojas, L. Z., Gómez-Ochoa, S. A., Verhoog, S., et al. (2022) · Figshare · doi:10.6084/m9.figshare.20522655
Roa‐Díaz, Z. M., Teuscher, J., Gamba, M., Bundo, M., Grisotto, G., Wehrli, F., Gamboa, E., Rojas, L. Z., Gómez-Ochoa, S. A., Verhoog, S., et al. (2022) · Figshare · doi:10.6084/m9.figshare.20522652
Nkwata, A. K., Song, X., Zhang, M., & Ezeamama, A. E. (2020) · Figshare · doi:10.6084/m9.figshare.12308603
Unpacking cancer diagnosis among Latin Americans in Canada: disparities and protective patterns
Escobar, K. M. (2025) · Canadian Journal of Public Health · doi:10.17269/s41997-025-01143-0
“Although some subgroup-specific odds ratios include 1.00 in their confidence intervals, the significance of the interaction terms indicates meaningful variation across subgroups (Greenland, 2009).”
InfluentialTruncated Fusion Learning on Supervised Clustering and Its Fast Stagewise Algorithm
Li, L., Li, Y., Zhang, J., & Zheng, Z. (2025) · INFORMS journal on computing · doi:10.1287/ijoc.2024.0840
Okubo, Y., Uda, K., Sawada, M., & Sakurai, Y. (2026) · Pediatric Pulmonology · doi:10.1002/ppul.71506
“Additive interactions were quantified using the relative excess risk due to interaction (RERI) [24], and multiplicative interactions were assessed by comparing the observed joint RR to the product of individual RRs [25].”
Wu, D., Lan, Y., Feng, B., Ding, X., Wu, C., Chen, S., Miao, C., Balmer, L. A., Li, X., Wu, S., et al. (2026) · The EPMA Journal · doi:10.1007/s13167-026-00439-6
Castellanos-Arreola, V., Castañeda-Márquez, A. C., Guerrero-Castañeda, R. F., Blanco-Hernández, D. M. R., Vázquez-García, L. R., Sánchez-Velázquez, J. C. F., Velázquez-Núñez, M. D. C., Castañeda-Ramírez, M. Y., & Hernández-Mariano, J. Á. (2026) · Clinics and Practice · doi:10.3390/clinpract16030049
“Additionally, to evaluate the robustness of the associations and to explore whether the relationship between NDs and mortality varied by clinical severity, we performed a sensitivity analysis assessing effect modification by the shock index [42,43].”
Beaudrap, P. D. (2009) · theses.fr (ABES) · doi:10.70675/858960b6z79f7z4862zbe28z0ad07cd497f3 · Cited by 1
Lalloué, B. (2013) · theses.fr (ABES) · doi:10.70675/bed1b010z8b1az43e6z87b2z26c5e0605631 · Cited by 1
Fu, J., Fei, F., Zhu, Y., Wu, K., Chen, X., Chen, F., Zhao, Q., Geng, M., Yao, Y., Yuan, X., et al. (2026) · Landscape and Urban Planning · doi:10.1016/j.landurbplan.2026.105666
Escobar, K. M. (2026) · Journal of Ethnic & Cultural Diversity in Social Work · doi:10.1080/15313204.2026.2675217
Metadata
- Authors
- Greenland, S.
- Affiliations
- University of California, Los Angeles (US)
- Corresponding author
- Greenland, Sander
- ORCID
- Greenland, Sander
- Venue
- Epidemiology
- Published
- 2009-01
- Volume
- 20
- Issue
- 1
- Pages
- 14-17
- ISSN
- 1044-3983 (ISSN-L)
- OpenAlex
- W1989872151
- Cited by
- 204
- References
- 30
- Citations by year
- 2026: 9 · 2025: 15 · 2024: 15 · 2023: 10 · 2022: 11 · 2021: 13 · 2020: 9 · 2019: 10 · 2018: 6 · 2017: 6 · 2016: 16 · 2015: 14 · 2014: 17 · 2013: 14 · 2012: 15
- FWCI
- 7.57
- 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
- Social Sciences › Arts and Humanities › History and Philosophy of Science › Philosophy of Biology and Scientific Explanation
- Keywords
- statistical interaction, synergistic effect, superadditivity, non-identifiability, intervention targeting
- MeSH terms
- Models, Theoretical, Terminology as Topic, Population Surveillance, Regression Analysis, Reproducibility of Results, Causality, Epidemiologic Studies, Risk Assessment
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.