Estimating interaction on an additive scale between continuous determinants in a logistic regression model
Open accessThe DOI matches its Crossref record; 7 of the 7 references it cites were found in a registry or an index.
Abstract
BACKGROUND: To determine the presence of interaction in epidemiologic research, typically a product term is added to the regression model. In linear regression, the regression coefficient of the product term reflects interaction as departure from additivity. However, in logistic regression it refers to interaction as departure from multiplicativity. Rothman has argued that interaction estimated as departure from additivity better reflects biologic interaction. So far, literature on estimating interaction on an additive scale using logistic regression only focused on dichotomous determinants. The objective of the present study was to provide the methods to estimate interaction between continuous determinants and to illustrate these methods with a clinical example. METHODS: and results From the existing literature we derived the formulas to quantify interaction as departure from additivity between one continuous and one dichotomous determinant and between two continuous determinants using logistic regression. Bootstrapping was used to calculate the corresponding confidence intervals. To illustrate the theory with an empirical example, data from the Utrecht Health Project were used, with age and body mass index as risk factors for elevated diastolic blood pressure. CONCLUSIONS: The methods and formulas presented in this article are intended to assist epidemiologists to calculate interaction on an additive scale between two variables on a certain outcome. The proposed methods are included in a spreadsheet which is freely available at: http://www.juliuscenter.nl/additive-interaction.xls.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
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?
The cited source says what the claim attributes to it.
In the citing paper
“This same tool could also be used in conjunction with a Cox proportional hazards model to produce measures of additive interaction,23,24 and to calculate additive interaction between continuous exposures.25”
From Recommendations for presenting analyses of effect modification and interaction (2012) · cites this paper as [25]
In the citing paper
“The 95% CI for RERI can be calculated using the delta method 22 or using the first percentile Bootstrap method.23 The latter is more suitable for continuous factors. 15”
From Understanding of interaction (subgroup) analysis in clinical trials (2019) · cites this paper as [15]
In the citing paper
“Also screened were interactions between traits of the big 5 with one another and with gender, age, and SES on both multiplicative and additive scales (36).”
From Personality, Socioeconomic Status, and All-Cause Mortality in the United States (2010) · cites this paper as [36]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Dichotomizing continuous predictors in multiple regression: a bad idea
Royston, P., Altman, D. G., & Sauerbrei, W. (2006) · Statistics in Medicine · doi:10.1002/sim.2331 · Cited by 2,312 · cited 1×
“Rothman called this amount of interaction RERI.2 Rewriting Formula ( 5 ) gives:”
Confidence interval estimation of interaction
Hosmer, D. W., & Lemeshow, S. (1992) · Epidemiology · doi:10.1097/00001648-199209000-00012 · Cited by 1,239 · cited 1×
“3 These studies only focused on interaction between two dichotomous determinants.”
Schottenfeld, D. (2002) · American Journal of Epidemiology · doi:10.1093/aje/kwf028 · Cited by 622
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 · cited 1×
“Note that the absolute background risk was 4.4% and thus the absolute risk due to interaction is 5.9% (4.4% � 1.34), which is exactly the amount of interaction for absolute risks calculated with Formula ( 4 ).”
InfluentialThe Utrecht Health Project: Optimization of routine healthcare data for research
Grobbee, D. E., Hoes, A. W., Verheij, T. J., Schrijvers, G., van Ameijden, E. J. C., & Numans, M. E. (2005) · European Journal of Epidemiology · doi:10.1007/s10654-004-5689-2 · Cited by 99
Self-citationConfidence interval estimation
Rees, D. G. (1989) · Springer eBooks · doi:10.1007/978-1-4899-7260-6_9 · Cited by 33
Rothman, K. J., Huybrechts, K. F., & Murray, E. J. (2024) · doi:10.1093/oso/9780197751541.001.0001 · Cited by 12
Cited by
Items 401–455 of 455
Luo, L., Xu, R., Mu, F., Li, H., Liu, Y., Gao, J., Wu, Y., Wang, K., Liu, Y., Zhang, Y., et al. (2024) · Journal of Affective Disorders · doi:10.1016/j.jad.2024.08.065
Hu, C., Lin, Z., Hu, Z., & Lin, S. (2024) · PeerJ · doi:10.7717/peerj.18304
“…concerning appropriate statistical methods for calculating the con fi dence interval of these measures ( Hosmer & Lemeshow, 1992 ; Assmann et al., 1996 ; Knol et al., 2007 ; Zou, 2008 ; Richardson & Kaufman, 2009 ; Kuss, Schmidt-Pokrzywniak & Stang, 2010 ; Nie et al., 2010 ; Chu, Nie & Cole, 2011 ).”
MethodJenssen, M., Arora, N., Løset, M., Åsvold, B. O., Thomas, L. F., Bekkevold, O.-J., Mai, X., Sun, Y., Furberg, A.-S., Jorde, R., et al. (2024) · JID Innovations · doi:10.1016/j.xjidi.2024.100336
Luo, P., Wang, F., Ying, J., Liu, K., Hua, B., Chen, S., Li, J., Sun, X., Ding, Y., Ye, B., et al. (2024) · Haematologica · doi:10.3324/haematol.2024.286328 · Cited by 2
“25,26 Mediation analysis examined air pollution’s role in greenspace and ITP associations.”
Additive and Multiplicative Interactions in Factorial Randomized Trials: What, Why, and How?
Mascha, E. J. (2024) · Anesthesia & Analgesia · doi:10.1213/ane.0000000000007175
Bathini, L., Jeyakumar, N., Sontrop, J. M., McArthur, E., Kang, Y., Luo, B., Bello, A. K., Collister, D., Ahmed, S. B., Kaul, P., et al. (2025) · Canadian Journal of Kidney Health and Disease · doi:10.1177/20543581251318836
Li, W., Yang, Y., Guo, S., Yi, J., Li, C. C., Shen, G., Wang, L., Chen, X., Zhi, Z., & Gao, H. N. (2025) · Social Psychiatry and Psychiatric Epidemiology · doi:10.1007/s00127-025-02901-8
“In the additive interaction analyses, the estimates derived from this model were subsequently employed to calculate the Relative Excess Risk due to Interaction (RERI) and the Attributable Proportion (AP) [ 20, 21 ].”
A Sex-specific Mendelian Randomization-Phenome-Wide Association Study of Body Mass IndexPreprint
Jiesisibieke, Z. L., Chan, I. I., Ng, J. C. M., & Schooling, C. M. (2025) · eLife · doi:10.7554/elife.102573.2
Cheng, H., Guo, P., Zhang, C., Zhang, G., Zhang, Z., Wang, S., Fida, S., & Song, C. (2025) · Research Square · doi:10.21203/rs.3.rs-6511788/v1
Tang, F. (2023) · Scholars Archive - University at Albany (University at Albany, State University of New York) · doi:10.54014/5c5p-5wbv
Jiesisibieke, Z. L., Chan, I. I., Ng, J. C. M., & Schooling, C. M. (2025) · eLife · doi:10.7554/elife.102573.3.sa2
Self‐Rated Health, Leisure Activities and Cognitive Frailty Among Community‐Dwelling Older Adults
Li, Y., Yang, J., Li, X., Jin, Y., Zhou, W., Yu, J., Chen, H., Wang, X., & Wang, C. (2025) · Psychogeriatrics · doi:10.1111/psyg.70056
“If the 95% CI of RERI and AP did not contain 0, or the 95% CI of S did not contain 1, there was additive interaction between the two variables [42].”
Huang, Z., Lu, J., He, G., Hu, J., Guo, X., Chen, M., Liu, T., Shao, L., Liu, F., Xu, Y., et al. (2025) · Environmental Research · doi:10.1016/j.envres.2025.122578
Islam, T., & Latif, A. H. M. M. (2025) · Epidemiologic Methods · doi:10.1515/em-2025-0007
Lin, Y., Shen, Y., Meng, X., Kan, H., Cai, J., & Hua, J. (2025) · The Lancet Regional Health - Western Pacific · doi:10.1016/j.lanwpc.2025.101701
“49 All statistical analyses were performed using R software (version 4.4.1, R Core Team 2024).”
InfluentialBox, C. V. J., van Oortmerssen, J. A. E., Tilly, M. J., Vernooij, M. W., Ikram, M. K., Kavousi, M., & Wolters, F. J. (2025) · Journal of the American Heart Association · doi:10.1161/jaha.125.042282
“29 We tested the proportional hazards assumptions with Schoenfeld residuals, combined with plots of the β coefficients over time (Figure S1).”
Ma, Q., Wu, Y., Huang, Y., Xie, M., Tao, S., Gu, Y., Liu, S., Warrier, V., & Wang, Q. (2025) · medRxiv · doi:10.1101/2025.11.10.25339885
Wu, S., Luo, H., Huang, R., Zhong, J., Su, M., Lai, X., Zhang, Z., & Zhou, Q. (2025) · European Journal of Nutrition · doi:10.1007/s00394-025-03841-4
“Additive interaction was evaluated by calculating the relative excess risk due to interaction (RERI), attributable proportion (AP), and synergy index (SI) [22].”
Arora, N., Bhatta, L., Skarpsno, E. S., Dalen, H., Åsvold, B. O., Brumpton, B. M., Richmond, R. C., & Strand, L. B. (2024) · Figshare · doi:10.6084/m9.figshare.24257634
Arora, N., Bhatta, L., Skarpsno, E. S., Dalen, H., Åsvold, B. O., Brumpton, B. M., Richmond, R. C., & Strand, L. B. (2023) · Figshare · doi:10.6084/m9.figshare.24257479
Tan, J.-Y., Zeng, Q.-L., Ni, M., Zhang, Y., & Qiu, T. (2022) · Figshare · doi:10.6084/m9.figshare.19625421
Lee, S., Lee, H. A., Park, B., Han, H., Hong, Y. S., Ha, E., & Park, H. (2024) · Figshare · doi:10.6084/m9.figshare.26559581
Fontvieille, E., Viallon, V., Recalde, M., Córdova, R., Jansana, A., Peruchet‐Noray, L., Lennon, H., Heath, A. K., Aune, D., Christakoudi, S., et al. (2023) · Figshare · doi:10.6084/m9.figshare.24628106
Ecological Weed Management for Field Crop Production
Menalled, U. (2023) · eCommons (Cornell University) · doi:10.7298/g7ph-h168
Zhang, C., Cui, J., Li, S., Shen, J., Luo, X., Yao, Y., & Shi, H. (2024) · Figshare · doi:10.6084/m9.figshare.25132123
Shi, H., Fu, J., Liu, X., Wang, Y., Yong, X., Jiang, L., Ma, S., Yin, Z., Yao, J., Yao, X., et al. (2021) · Figshare · doi:10.6084/m9.figshare.15178943
Neale, Z. E., Fonda, J. R., Miller, M. W., Wolf, E. J., Zhang, R., Sherva, R., Harrington, K. M., Merritt, V. C., Panizzon, M. S., Hauger, R. L., et al. (2024) · Figshare · doi:10.6084/m9.figshare.26740958
The Ethnic and Social Determinants of Dementia
Bothongo, P. L. (2024) · Queen Mary Research Online (Queen Mary University of London)
Benka-Coker, W. O., Magzamen, S., Peel, J., Wilson, A., & Anderson, B. (2019) · Digital Collections of Colorado (Colorado State University) · doi:10.25675/3.020656
Cao, L., Sundquist, K., Zhang, Y., Memon, A. A., Hedelius, A., Li, N., Ji, J., Sundquist, J., Zheng, D., & Wang, X. H. (2026) · Journal of Affective Disorders · doi:10.1016/j.jad.2026.121394
The Influence of Spousal Genotype on Alcohol Problems in Marriage
Kuo, S. I., McCutcheon, V. V., Alıev, F., Bucholz, K. K., Dick, D. M., Edenberg, H. J., HESSELBROCK, V. M., Kramer, J. R., Lai, D., Plawecki, M. H., et al. (2026) · Alcoholism Clinical and Experimental Research · doi:10.1111/acer.70399
“Because AUDcrit was analyzed as a continuous, log-transformed outcome using linear mixed-effects models, product terms estimate statistical interaction, or departure from additivity, on the transformed AUDcrit scale (Basten et al. 2025; Knol et al. 2007).”
Risk factors for suicidal ideation among bereaved adolescents in a psychological support hotline
Liu, Z.-K., Wang, R.-F., Li, X.-X., Yuan, Z., Wu, T. T., Zhao, L.-T., Chen, L., & Liang, H. (2026) · Scientific Reports · doi:10.1038/s41598-026-41739-1
Zoghbi, M. E. (2016) · theses.fr (ABES) · doi:10.70675/36569deez9a94z4f09z9cb7za9e977396e63
Lu, Z.-A., Ploner, A., & Bergen, S. E. (2026) · Psychological Medicine · doi:10.1017/s0033291726104012
“…the sum of each effect alone), whereas a RERI with a discordant sign with OR High PRS + ACEs indicates the presence of an additive antagonistic effect (combined effects smaller than the sum of each effect alone) (Knol, van der Tweel, Grobbee, Numans, & Geerlings, 2007; Knol & VanderWeele, 2012).”
Li, L., Xiao, S., Zhang, Z., Yang, H., & Tao, Y. (2026) · Frontiers in Nutrition · doi:10.3389/fnut.2026.1744184
“Additive interaction effects were evaluated using the relative excess risk due to interaction (RERI), attributable proportion due to interaction (AP), and synergy index (S), and their 95% confidence interval (CIs) and p values were calculated using the delta method (24, 25).”
L'impact de l'adiposité et des maladies cardiométaboliques sur l'incidence et la survie du cancer
Fontvieille, E. (2024) · theses.fr (ABES) · doi:10.70675/9e8c93c3z732fz4707zb3d9z03df95400733
Bayarmunkh, O., Ganbold, C., Unurbat, E., Byambadorj, B., Battulga, Z., Budragchaa, D., Battulga, P., Janakhmyed, M., Davaatseren, U., & Jav, S. (2026) · Ophthalmic Genetics · doi:10.1080/13816810.2026.2657423
Contribution of FXR1 genetic polymorphisms to breast cancer susceptibility in Chinese females
Zhang, C., Zhang, G., Fida, S., Zhang, Y., Wang, S., Zhang, Z. J., & Song, C. (2026) · Molecular Genetics and Genomics · doi:10.1007/s00438-026-02436-w
“No significant additive interaction was determined if the 95% CI of S included 1, or if those of AP and RERI included 0.”
Bjørndal, L. D., Hegemann, L., Pettersen, J. H., Corfield, E. C., Hannigan, L. J., Yılmaz, Z., Costello, K., Ystrøm, E., Reichborn‐Kjennerud, T., Andreassen, O. A., et al. (2026) · Biological Psychiatry Global Open Science · doi:10.1016/j.bpsgos.2026.100761
“In exploratory analyses, we examined additive interaction effects by quantifying the relative excess risk due to interaction (RERI) with the following equation (46): The 95% CIs for RERI estimates were obtained by boot-strapping ( n = 1000).”
Verkuilen, A., Talbert, L. D., Valocchi, E., Kodama, L., Coffinier, C., Program, V. A. M. V., & Merritt, V. C. (2026) · Journal of Neurotrauma · doi:10.1177/08977151261450343
Clark, A. L., Valocchi, E., McGill, M. B., Lopez, F. V., Coffinier, C., Thomas, K. R., Jester, D. J., Logue, M. W., & Merritt, V. C. (2026) · Alzheimer s Research & Therapy · doi:10.1186/s13195-026-02086-5
“We also tested for potential additive interactions of TBI and the set of health conditions and health behaviors by calculating the relative excess risk due to interaction (RERI) statistic, adjusting for age, sex, and education [29].”
Per- and polyfluoroalkyl substances, vitamin D, and risk of preterm birth in a Canadian cohort
Fisher, M. M., Owen, J., Marro, L., Potter, B., Weiler, H. A., England-Mason, G., Borghese, M. M., Walker, M., DODDS, L., Helewa, M. E., et al. (2026) · Environmental Research · doi:10.1016/j.envres.2026.124999
Yang, Z., Ng, C. F. S., Ueda, K., Lee, W., Stafoggia, M., Araki, S., Shimadera, H., Chatani, S., Kojima, S., Ikeda, T., et al. (2026) · Environment & Health · doi:10.1021/envhealth.6c00094
Błaszczyk, R. T., Sawonik, S., Korona‐Głowniak, I., Wysocka, A., Czuba, M., Świstowska, M., KRÓL, O., Kocki, J., Wysokiński, A., & Główniak, A. (2026) · International Journal of Molecular Sciences · doi:10.3390/ijms27156710
“The authors observed an increased risk of atrial fibrillation recurrence in carriers of both the 5A and 1G alleles compared to people without these alleles [19,20].”
Pain Phenotypes and Falls in Older Adults: Cross-Sectional Associations by Social Engagement
Zhao, R., Feng, Q., Yu, P., Li, W., & Gu, Y. (2026) · Pain Management Nursing · doi:10.1016/j.pmn.2026.07.003
Lu, S.-Y., Xu, X.-D., Wang, D.-Q., Amjad, M., Dong, G., Lin, L.-Z., & Hong-wei, T. U. (2026) · Atmospheric Environment · doi:10.1016/j.atmosenv.2026.122270
Stressful events, childhood adversities, cortisol activity and anxiety in older adults
Bakouni, H. (2021) · Knowledge UdeS (Institutional Deposit of the University of Sherbrooke)
Duan, Y., Wang, Y., Li, F., Li, W., Zhang, D., & Cui, Y. (2026) · Biological Trace Element Research · doi:10.1007/s12011-026-05302-3
“Different statistical models may yield inconsistent results, whereas biological interactions are pre - ferred in etiologic research [ 54].”
Sun, Y., Yu, Y., Zhu, C., Cai, L., Tan, X., Lu, Y., Wang, N., Wang, Y., & Xia, F. (2026) · npj Aging · doi:10.1038/s41514-026-00497-0
Santiago, A. C., Ahmed, A. E., Yoon, J., & Alcover, K. C. (2026) · Military Medicine · doi:10.1093/milmed/usag388
Duan, Y., Wang, Y., Li, F., Li, W., Zhang, D., & Cui, Y. (2026) · Nutrients · doi:10.3390/nu18183085
“Multiplicative and additive models were applied to examine interactions between influencing factors and two polymorphisms, TSHR rs2268458 and NKX2-1 rs2076735 [45,46].”
Interaction between adiposity and genetic risk score on blood pressure traits in Mexican adults
Argoty-Pantoja, A. D., Velázquez‐Cruz, R., Aparicio-Trujillo, E., Bonilla, F., Torres-Hurtado, G., Cortés-García, S. P., Snieder, H., Rivera‐Paredez, B., & Salmerón, J. (2026) · Journal of Human Hypertension · doi:10.1038/s41371-026-01222-w
Chen, Y., Zhao, T., Li, C., Meng, H., & Qiao, H. (2026) · Frontiers in Digital Health · doi:10.3389/fdgth.2026.1920279
“The additive interaction model, which is grounded in the risk difference scale that quantifies absolute changes in risk, aligns more closely with the biological concept of synergy and is of greater relevance to public health (57).”
Metadata
- Authors
- Knol, M. J., van der Tweel, I., Grobbee, D. E., Numans, M. E., & Geerlings, M. I.
- Affiliations
- Utrecht University (NL)
- University Medical Center Utrecht (NL)
- Corresponding author
- Knol, Mirjam J
- Venue
- International Journal of Epidemiology
- Publisher
- Oxford University Press (OUP)
- Published
- 2007-08-27
- Volume
- 36
- Issue
- 5
- Pages
- 1111-1118
- ISSN
- 0300-5771 (ISSN-L)
- 1464-3685
- OpenAlex
- W2114552460
- Cited by
- 444
- References
- 7
- Citations by year
- 2026: 27 · 2025: 39 · 2024: 46 · 2023: 40 · 2022: 43 · 2021: 23 · 2020: 22 · 2019: 18 · 2018: 23 · 2017: 11 · 2016: 21 · 2015: 24 · 2014: 27 · 2013: 18 · 2012: 16
- FWCI
- 8.21
- Type
- journal article
- Language
- en
- Open access
- bronze
- doi.org
- Locations
- International Journal of Epidemiology (publishedVersion)
- PubMed (publishedVersion)
- Utrecht University Repository (Utrecht University) (submittedVersion, other-oa)
- Funders
- Novo Nordisk
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Epidemiology
- Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Genetics › Genetic Associations and Epidemiology
- Keywords
- interaction estimation, logistic regression, interaction terms, multiplicative interaction, bootstrapping, confidence intervals, diastolic blood pressure, body mass index, age
- MeSH terms
- Adolescent, Adult, Age Factors, Aged, Aged, 80 and over, Child, Data Interpretation, Statistical, Humans, Hypertension, Middle Aged, Netherlands, Body Mass Index, Logistic Models
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.