Authors often do not give sufficient information to draw conclusions about the size and statistical significance of interaction on the additive and multiplicative scales. To improve this, we provide ...four steps, template tables and examples. We distinguish two cases: when the causal effect of intervening on one exposure, across strata of another factor, is of interest ('effect modification'); and when the causal effect of intervening on two exposures is of interest ('interaction'). Assume we study whether X modifies the effect of A on D, where A, X and D are dichotomous. We propose presenting: (i) relative risks (RRs), odds ratios (ORs) or risk differences (RDs) for each (A, X) stratum with a single reference category taken as the stratum with the lowest risk of D; (ii) RRs, ORs or RDs for A within strata of X; (iii) interaction measures on additive and multiplicative scales; (iv) the A-D confounders adjusted for. Assume we study the interaction between A and B on D, where A, B and D are dichotomous. Steps (i) and (iii) are similar to presenting effect modification. (ii) Present RRs, ORs or RDs for A within strata of B and for B within strata of A. (iv) List the A-D and B-D confounders adjusted for. These four pieces of information will provide a reader the information needed to assess effect modification or interaction. The presentation can be further enriched when exposures have multiple categories. Our proposal hopefully encourages researchers to present effect modification and interaction analyses in as informative a manner as possible.
A Tutorial on Interaction VanderWeele, Tyler J.; Knol, Mirjam J.
Epidemiologic methods,
01/2014, Letnik:
3, Številka:
1
Journal Article
Recenzirano
Odprti dostop
In this tutorial, we provide a broad introduction to the topic of interaction between the effects of exposures. We discuss interaction on both additive and multiplicative scales using risks, and we ...discuss their relation to statistical models (e.g. linear, log-linear, and logistic models). We discuss and evaluate arguments that have been made for using additive or multiplicative scales to assess interaction. We further discuss approaches to presenting interaction analyses, different mechanistic forms of interaction, when interaction is robust to unmeasured confounding, interaction for continuous outcomes, qualitative or “crossover” interactions, methods for attributing effects to interactions, case-only estimators of interaction, and power and sample size calculations for additive and multiplicative interaction.
Abstract
Given the emergence of the SARS-CoV-2 Omicron BA.1 and BA.2 variants and the roll-out of booster COVID-19 vaccination, evidence is needed on protection conferred by primary vaccination, ...booster vaccination and previous SARS-CoV-2 infection by variant. We employed a test-negative design on S-gene target failure data from community PCR testing in the Netherlands from 22 November 2021 to 31 March 2022 (n = 671,763). Previous infection, primary vaccination or both protected well against Delta infection. Protection against Omicron BA.1 infection was much lower compared to Delta. Protection was similar against Omicron BA.1 compared to BA.2 infection after previous infection, primary and booster vaccination. Higher protection was observed against all variants in individuals with both vaccination and previous infection compared with either one. Protection against all variants decreased over time since last vaccination or infection. We found that primary vaccination with current COVID-19 vaccines and previous SARS-CoV-2 infections offered low protection against Omicron BA.1 and BA.2 infection. Booster vaccination considerably increased protection against Omicron infection, but decreased rapidly after vaccination.
In many countries, serotype 19A became the most frequently carried serotype in children and the dominant cause of invasive pneumococcal disease in children and other age groups.2–4 Moreover, serotype ...19A is often associated with antibiotic resistance.4 Although serotype 19A is not included in PCV10, the vaccine was licensed for the prevention of 19A invasive pneumococcal disease, based on immunogenicity data and vaccine effectiveness data as observed following PCV10 introduction. Data from Brazil and Canada reported an effectiveness of 71–82% for PCV10 against serotype 19A invasive pneumococcal disease.5,6 By contrast, in Finland there was no significant effectiveness of PCV10, and in Sweden there were increases in 19A invasive pneumococcal disease among children in counties vaccinating with PCV10.7,8 In the Netherlands, PCV10 had no effect on carriage of serotype 19A in a randomised controlled trial comparing PCV7 with PCV10.9 Although the incidence of serotype 19A invasive pneumococcal disease in children younger than 5 years has been stable in the Netherlands since the PCV10 introduction, with approximately ten cases per year, serotype 19A has remained a dominant serotype in carriage and invasive pneumococcal disease.10–12 So, there is still uncertainty about the extent and duration of cross-protection of PCV10 against serotype 19A in children. AvdE has received a grant from Pfizer for research on pneumococcal infections (Investigator Initiated project “Epidemiology of invasive pneumococcal disease”, IIR WI173197), participated in advisory boards of Pfizer, and does consultancy activities for GlaxoSmithKline (fees paid to Amsterdam University Medical Center).
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.
Measures of interaction on an additive scale (relative excess risk due to interaction RERI, attributable proportion AP, synergy index S), were developed for risk factors rather than preventive ...factors. It has been suggested that preventive factors should be recoded to risk factors before calculating these measures. We aimed to show that these measures are problematic with preventive factors prior to recoding, and to clarify the recoding method to be used to circumvent these problems. Recoding of preventive factors should be done such that the stratum with the lowest risk becomes the reference category when both factors are considered jointly (rather than one at a time). We used data from a case-control study on the interaction between ACE inhibitors and the ACE gene on incident diabetes. Use of ACE inhibitors was a preventive factor and DD ACE genotype was a risk factor. Before recoding, the RERI, AP and S showed inconsistent results (RERI = 0.26 95% CI: -0.30; 0.82, AP = 0.30 95% CI: -0.28; 0.88, S = 0.35 95% CI: 0.02; 7.38), with the first two measures suggesting positive interaction and the third negative interaction. After recoding the use of ACE inhibitors, they showed consistent results (RERI = -0.37 95% CI: -1.23; 0.49, AP = -0.29 95% CI: -0.98; 0.40, S = 0.43 95% CI: 0.07; 2.60), all indicating negative interaction. Preventive factors should not be used to calculate measures of interaction on an additive scale without recoding.
The 7-valent pneumococcal conjugate vaccine (PCV7) was introduced in The Netherlands in 2006 and was replaced by PHiD-CV10 in 2011. Data on carriage prevalence of S. pneumoniae serotypes in children ...and invasive pneumococcal disease (IPD) in children and older adults were collected to examine the impact of PCVs on carriage and IPD in The Netherlands. Pneumococcal carriage prevalence was determined by conventional culture of nasopharyngeal swabs in 24-month-old children in 2015/2016. Data were compared to similar carriage studies in 2005 (pre-PCV7 introduction), 2009, 2010/2011 and 2012/2013. Invasive pneumococcal disease isolates from hospitalized children <5 years and adults >65 years (2004-2016) were obtained by sentinel surveillance. All isolates were serotyped by Quellung. Serotype invasive disease potential was calculated using carriage and nationwide IPD data in children. The overall pneumococcal carriage rate was 48% in 2015/2016, lower than in 2010/2011 (64%) and pre-vaccination in 2005 (66%). Carriage of the previously dominant non-vaccine serotypes 19A and 11A has declined since 2010/2011, from 14.2% to 4.6% and 4.2% to 2.7%, respectively, whereas carriage of serotypes 6C and 23B has increased (4.2% to 6.7% and 3.9% to 7.3%), making serotypes 6C and 23B the most prevalent carriage serotypes. IPD incidence declined in children (20/100,000 cases in 2004/2006 to 6/100,000 cases in 2015/2016) as well as in older adults (63/100,000 cases to 51/100,000 cases). Serotypes 6C, 23B and 11A have high carriage prevalence in children, but show low invasive disease potential. Serotype 8 is the main causative agent for IPD in older adults (11.3%). In conclusion, 10 years after the introduction of pneumococcal vaccination in children in The Netherlands shifts in carriage and disease serotypes are still ongoing. Surveillance of both carriage and IPD is important to assess PCV impact and to predict necessary future vaccination strategies in both children and older adults.
Celotno besedilo
Dostopno za:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
An increasing proportion of the population has acquired immunity through COVID-19 vaccination and previous SARS-CoV-2 infection, i.e., hybrid immunity, possibly affecting the risk of new infection. ...We aim to estimate the protective effect of previous infections and vaccinations on SARS-CoV-2 Omicron infection, using data from 43,257 adult participants in a prospective community-based cohort study in the Netherlands, collected between 10 January 2022 and 1 September 2022. Our results show that, for participants with 2, 3 or 4 prior immunizing events (vaccination or previous infection), hybrid immunity is more protective against infection with SARS-CoV-2 Omicron than vaccine-induced immunity, up to at least 30 weeks after the last immunizing event. Differences in risk of infection are partly explained by differences in anti-Spike RBD (S) antibody concentration, which is associated with risk of infection in a dose-response manner. Among participants with hybrid immunity, with one previous pre-Omicron infection, we do not observe a relevant difference in risk of Omicron infection by sequence of vaccination(s) and infection. Additional immunizing events increase the protection against infection, but not above the level of the first weeks after the previous event.
Logistic regression analysis, which estimates odds ratios, is often used to adjust for covariables in cohort studies and randomized controlled trials (RCTs) that study a dichotomous outcome. In ...case-control studies, the odds ratio is the appropriate effect estimate, and the odds ratio can sometimes be interpreted as a risk ratio or rate ratio depending on the sampling method.1-4 However, in cohort studies and RCTs, odds ratios are often interpreted as risk ratios. This is problematic because an odds ratio always overestimates the risk ratio, and this overestimation becomes larger with increasing incidence of the outcome.5 There are alternatives for logistic regression to obtain adjusted risk ratios, for example, the approximate adjustment method proposed by Zhang and Yu5 and regression models that directly estimate risk ratios (also called "relative risk regression").6-9 Some of these methods have been compared in simulation studies.7,9 The method by Zhang and Yu has been strongly criticized,7,10 but regression models that directly estimate risk ratios are rarely applied in practice. We found eight methods to estimate adjusted risk ratios in the literature (Table 35,7-9,14-19). The Mantel-Haenszel risk ratio method is straightforward and gives a weighted risk ratio over strata of covariables.14,15 This method is only practica- ble if adjusting for a small number of categorical covariables (i.e., continuous covariables first need to be categorized). Log-binomial and Poisson regression are generalized linear models that directly estimate risk ratios.7,8 The default standard errors obtained by Poisson regression are typically too large; therefore, calculation of robust standard errors for Poisson regression may be needed to obtain a correct confidence interval around the risk ratio.9 The other four methods use odds ratios or logistic regression to estimate risk ratios. The Zhang and Yu method is a simple formula that calculates the risk ratio based on the odds ratio and the incidence of the outcome in the unexposed group.5 The doublingof- cases method concerns changing the data set in such a way that logistic regression yields a risk ratio instead of an odds ratio.17 Again, calculation of robust standard errors may be needed to obtain a correct confidence interval around the risk ratio.18 Lastly, the method proposed by Austin uses the predicted probabilities obtained from a logistic regression model to estimate risk ratios.19 A recent review article of methods to estimate risk ratios and risk differences in cohort studies illustrated several of these eight methods using empirical data.20 We showed in the clinical examples and simulations that an odds ratio can substantially overestimate the risk ratio. In fact, both are correct, but when an odds ratio is interpreted as a risk ratio, serious misinterpretation with potential consequences for treatment decisions and policymaking can occur, as illustrated by the two clinical examples. Therefore, any misinterpretation of odds ratios should be avoided with calculation and presentation of adjusted risk ratios in both cohort studies and RCTs. Also, if adjustment for baseline covariates is not done, which is often the case in RCTs, the risk ratio is the preferred measure of association in case of dichotomous outcomes.21 Note that in case-control studies, the odds ratio is the appropriate effect estimate and the odds ratio can be interpreted as a risk ratio or rate ratio depending on the sampling method.1-4 Of course, if data of cohort studies or RCTs are collected so that a time-dependent analysis is possible, Cox regression yielding hazard ratios is recommended because it estimates relative hazards and does not involve problems related to odds ratios.
PDF
