Relative Risk Scale Research Articles

Simultaneous evaluation of the imprecision and inconsistency domains of GRADE can be performed using prediction intervals

ObjectivesTo explore the use of prediction interval (PI) for the simultaneous evaluation of the imprecision and inconsistency domains of Grading of Recommendations, Assessment, and Evaluation using stakeholder-provided decision thresholds. Study Design and SettingWe propose transforming the PI of a meta-analysis from a relative risk scale to an absolute risk difference using an appropriate baseline risk. The transformed PI is compared to stakeholder-provided thresholds on an absolute scale. We applied this approach to a large convenience sample of meta-analyses extracted from the Cochrane Database of Systematic Reviews and compared it against the traditional approach of rating imprecision and inconsistency separately using confidence intervals and statistical measures of heterogeneity, respectively. We used empirically derived thresholds following Grading of Recommendations, Assessment, and Evaluation guidance. ResultsThe convenience sample consisted of 2516 meta-analyses (median of 7 studies per meta-analysis; interquartile range: 5–11). The main analysis showed the percentage of meta-analyses in which both approaches had the same number of certainty levels rated down was 59%. The PI approach led to more levels of rating down (lower certainty) in 27% and to fewer levels of rating down (higher certainty) in 14%. Multiple sensitivity analyses using different thresholds showed similar results, but the PI approach had particularly increased width with a larger number of included studies and higher I2 values. ConclusionUsing the PI for simultaneous evaluation of imprecision and inconsistency seems feasible and logical but can lead to lower certainty ratings. The PI-based approach requires further testing in future systematic reviews and guidelines using context-specific thresholds and evidence-to-decision criteria. Plain Language SummaryThe prediction interval (PI) addresses both the imprecision and inconsistency domains of certainty. In this study, we applied this PI approach to simultaneously judge both domains and compared this to the traditional approach of making these separate judgments. The 2 approaches had moderate agreement. The PI-based approach requires further testing in future systematic reviews and guidelines using context-specific thresholds and evidence-to-decision criteria.

Amino Acid Infusion for Perioperative Functional Renal Protection: A Meta-analysis

ObjectiveAcute kidney injury (AKI) is a common perioperative complication. To date, no single intervention has been proven effective for AKI prevention in this setting. However, intravenous amino acids (AA) administration may recruit renal functional reserve and, thereby, attenuate the perioperative loss of glomerular filtration rate. Accordingly, we performed a meta-analysis to assess the efficacy of AA infusion for perioperative renal functional protection. MethodsWe performed a meta-analysis of controlled studies in perioperative patients (P) evaluating intravenous AA infusion (I) versus any comparator (C). The primary outcome was AKI at longest follow-up (O). We performed a random effects meta-analysis on the relative risk (RR) scale to assess the effect of AA infusion. We used a Bayesian approach to estimate the probability of benefit (RR<1) for the primary outcome. Secondary outcomes included renal-replacement therapy, serum creatinine, and estimated glomerular filtration rate. Tertiary outcomes included mechanical ventilation duration, intensive care unit and hospital length of stay and mortality. (PROSPERO: CRD42024547225). ResultsWe identified 15 studies (14 RCTs and one prospective before-after study) reporting at least one outcome of interest (4,544 patients), with six studies (4,084 patients) reporting the primary outcome. AKI occurred 504/2,041 (24.7%) in AA patients vs 614/2,041 (30.1%) in controls (RR=0.66; 95% CI, 0.47 to 0.94; I2=50%; p=0.02), which corresponded to a 99.1% probability of AKI reduction with AA. Moreover, coherent with such findings, AA reduced serum creatinine, and hospital length of stay, and increased eGFR. ConclusionsThis meta-analysis suggests that AA administration likely reduces the perioperative incidence of AKI.

Choosing Statistical Models to Assess Biological Interaction as a Departure from Additivity of Effects

Vanderweele and Knol define biological interaction as an instance wherein “two exposures physically interact to bring about the outcome.” A hallmark of biological interaction is that the total effect, produced when factors act together, differs from the sum of effects when the factors operate independently. Epidemiologists construct statistical models to assess biological interaction. The form of the statistical model determines whether it is suited to detecting departures from additivity of effects or for detecting departures from multiplicativity of effects. A consensus exists that biological interaction should be assessed as a departure from additivity of effects.This paper compares three statistical models’ assessment of a data example that appears in several epidemiology textbooks to illustrate biological interaction in a binomial outcome. A linear binomial model quantifies departure from additivity in the data example in terms of differences in probabilities. It generates directly interpretable estimates and 95% confidence intervals for parameters including the interaction contrast (IC). Log binomial and logistic regression models detect no departure from multiplicativity in the data example. However, their estimates contribute to calculation of a “Relative Excess Risk Due to Interaction” (RERI), a measure of departure from additivity on a relative risk scale. The linear binomial model directly produces interpretable assessments of departures from additivity of effects and deserves wider use in research and in the teaching of epidemiology. Strategies exist to address the model’s limitations.

Application of multi-criteria decision analysis techniques and decision support framework for informing plant select agent designation and decision making.

The United States Department of Agriculture (USDA) Division of Agricultural Select Agents and Toxins (DASAT) established a list of biological agents (Select Agents List) that threaten crops of economic importance to the United States and regulates the procedures governing containment, incident response, and the security of entities working with them. Every 2years the USDA DASAT reviews their select agent list, utilizing assessments by subject matter experts (SMEs) to rank the agents. We explored the applicability of multi-criteria decision analysis (MCDA) techniques and a decision support framework (DSF) to support the USDA DASAT biennial review process. The evaluation includes both current and non-select agents to provide a robust assessment. We initially conducted a literature review of 16 pathogens against 9 criteria for assessing plant health and bioterrorism risk and documented the findings to support this analysis. Technical review of published data and associated scoring recommendations by pathogen-specific SMEs was found to be critical for ensuring accuracy. Scoring criteria were adopted to ensure consistency. The MCDA supported the expectation that select agents would rank high on the relative risk scale when considering the agricultural consequences of a bioterrorism attack; however, application of analytical thresholds as a basis for designating select agents led to some exceptions to current designations. A second analytical approach used agent-specific data to designate key criteria in a DSF logic tree format to identify pathogens of low concern that can be ruled out for further consideration as select agents. Both the MCDA and DSF approaches arrived at similar conclusions, suggesting the value of employing the two analytical approaches to add robustness for decision making.

Application of multi-criteria decision analysis techniques and decision support framework for informing select agent designation for agricultural animal pathogens.

The United States Department of Agriculture (USDA), Division of Agricultural Select Agents and Toxins (DASAT) established a list of biological agents and toxins (Select Agent List) that potentially threaten agricultural health and safety, the procedures governing the transfer of those agents, and training requirements for entities working with them. Every 2years the USDA DASAT reviews the Select Agent List, using subject matter experts (SMEs) to perform an assessment and rank the agents. To assist the USDA DASAT biennial review process, we explored the applicability of multi-criteria decision analysis (MCDA) techniques and a Decision Support Framework (DSF) in a logic tree format to identify pathogens for consideration as select agents, applying the approach broadly to include non-select agents to evaluate its robustness and generality. We conducted a literature review of 41 pathogens against 21 criteria for assessing agricultural threat, economic impact, and bioterrorism risk and documented the findings to support this assessment. The most prominent data gaps were those for aerosol stability and animal infectious dose by inhalation and ingestion routes. Technical review of published data and associated scoring recommendations by pathogen-specific SMEs was found to be critical for accuracy, particularly for pathogens with very few known cases, or where proxy data (e.g., from animal models or similar organisms) were used to address data gaps. The MCDA analysis supported the intuitive sense that select agents should rank high on the relative risk scale when considering agricultural health consequences of a bioterrorism attack. However, comparing select agents with non-select agents indicated that there was not a clean break in scores to suggest thresholds for designating select agents, requiring subject matter expertise collectively to establish which analytical results were in good agreement to support the intended purpose in designating select agents. The DSF utilized a logic tree approach to identify pathogens that are of sufficiently low concern that they can be ruled out from consideration as a select agent. In contrast to the MCDA approach, the DSF rules out a pathogen if it fails to meet even one criteria threshold. Both the MCDA and DSF approaches arrived at similar conclusions, suggesting the value of employing the two analytical approaches to add robustness for decision making.

THE RESULTS OF THE ANALYSIS OF RISK FACTORS, CARDIOVASCULAR RISK, CHRONIC NON-COMMUNICABLE DISEASES IN YOUNG AND MIDDLE-AGED PATIENTS, DEPENDING ON THE EDUCATION

To assess the frequency of risk factors (RF), total cardiovascular risk (CHR) and their association with the level of education in young and middle-aged people. Persons aged 25-59 (40.4 ± 9.2) years old took part in a one-stage comparative study; an anamnesis was taken, a physical examination was carried out, risk factors for cardiovascular diseases, cardiovascular risk were assessed according to the Systematic Coronary Risk Estimation scale in persons aged 40 years and older, according to the relative risk scale - under the age of 40, laboratory parameters. Hypercholesterolemia was present in 6.1%, hyperglycemia - in 4%, obesity - in 2.5%, one risk factor - in 25.5%, 3 or more risk factors - in 30.7% of patients, the maximum number of risk factors - in men. Moderate CVR was present in 58.4%, very high - in 3.7%, low - in 31.8% of cases. Low CV risk: prevalence in women with higher education (p = 0.034), compared with women with secondary education, which was not observed in the group of men (p = 0.109). Men smoked more. Persons with higher education quit smoking 4 times more often than persons with secondary education (p = 0.001; OR = 3.98), persons with secondary education smoked 2.74 times more often (p = 0.001, OR = 2.74), than higher education. Overweight was detected in 47%: in females (p < 0.001) and males (p = 0.003), its occurrence was less common in the group with higher education. AH was present in 8.3% of patients, and the level of its control was better in those with higher education than those with secondary education (p < 0.001). The gender conditionality of low CVR and unidirectional trends in attitudes towards smoking in connection with the level of education were established. Purposeful consideration of the level of education, age, gender, increases the identification of risk groups for the formation of multimorbidity and high cardiovascular risk in the future.

Implementing an Artificial Intelligence System in the Work of General Practitioner in the Yamalo-Nenets Autonomous Okrug: Pilot Cross-sectional Screening Observational Study

Background. Early identification of risk factors (RF) associated with cardiovascular diseases (CVD) is essential for the prevention of CVDs and their complications. CVD risk factors can be identified using Artificial Intelligence (AI) systems, which are capable of learning, analyzing and drawing conclusions. The advantage of AI systems consists in their capacity to process large amounts of data over a short period of time and produce ready-made information. Objectives. Evaluation of the efficiency of implementing an AI software application by a general practitioner for identifying CVD risk factors.Methods. The study included data from 1778 electronic medical histories of patients aged over 18, assigned to an outpatient and polyclinic department of Muravlenkovskaya Gorodskaya Bolnitsa (Muravlenko municipal hospital), Yamalo-Nenets Autonomous Okrug (Russia). The study was conducted in four stages. The first stage involved a preliminary training of the Artificial Intelligence (AI) system under study using numerous CVD risk assessment scales. The Webiomed predictive analytics and risk management software by K-SkAI, Russia, was selected as a platform for this purpose. The second stage included an analysis of medical data to identify CVD risk factors according to the relative risk scale for patients under 40 and the SCORE scale for patients over 40. At the third stage, a specialist analyzed the previous and new information received about each patient. According to the results of the third stage, four risk groups for CVD (low, medium, high and very high) were formed. At the fourth stage, newly diagnosed patients with a high risk of CVD, who had not been previously subject to regular medical check-up, were directed for additional clinical, laboratory and instrumental follow-up examination and consultations of relevant specialists. Statistical data in absolute terms and as a percentage were obtained. Statistical processing of the results was carried out by a computer program aimed at medical decision support. Content visualization was performed in spreadsheets and charts.Results. Based on the data obtained, the AI system under study divided all patients into CVD risk groups and identified uncounted factors. The AI system confirmed a high and very high risk of CVD according to SCORE (Systematic COronary Risk Evaluation) in 623 people, who were already receiving appropriate cardiological assistance. The RFs that had not previously been taken into account in the diagnosis were recorded in 41 (11.5%) patients from the very highrisk group and in 37 (12.7%) high-risk patients. The AI system identified a high risk of CVD in 29 people who had not been previously under care of a general practitioner or other specialists due to their infrequent visits to health care facilities. These patients were detected by the AI system following periodic and preliminary medical check-ups (35%), full in-patient treatment for other diseases (31%), when seeking help of other specialists (17%), as well as when obtaining a medical certificate for a driving license (12%), admission to a swimming pool (3%) or possessing a weapon (2%). In a group with the newly diagnosed patients at a high risk of CVD, men dominated (24 persons, 82%) and women comprised only 8% (5 persons). All these people were of working age between 40 and 50. In order to confirm the information received, the supervising physician subsequently referred patients for a follow-up examination, as a result of which only 1 person (3%) was not diagnosed with a somatic pathology.Conclusion. The efficiency of the AI system under study comprised 97%. Permanent monitoring of all parameters of electronic medical histories and outpatient records is an efficient method for timely identification of RF at any visit of a person to a health care facility (preventive and periodic medical examinations, regular check-ups, specialist consultations, etc.) and their assignment to respective CVD risk groups. Such monitoring ensures an effective medical supervision of able-bodied populations.

Sex differences in associations of comorbidities with incident cardiovascular disease: focus on absolute risk.

AimTo examine sex differences in associations of obesity, type-2 diabetes, hypertension, and atrial fibrillation (AF) with incident cardiovascular disease (CVD), focusing on absolute risk measures.Methods and resultsWe included a total of 7994 individuals (mean age 49.1 years; 51.2% women) without prior CVD from the PREVEND (Prevention of Renal and Vascular End-stage Disease) cohort with a median follow-up of 12.5 years. Using Poisson regression, we calculated the increase in absolute as well as relative CVD risk associated with a comorbidity using incidence rate differences (IRD = IRcomorbidity−IRno-comorbidity) and incidence rate ratios (IRR = IRcomorbidity/IRno-comorbidity), respectively. Sex differences were presented as women-to-men differences (WMD = IRDwomen−IRDmen) and women-to-men ratios (WMR = IRRwomen/IRRmen). Absolute CVD risk was lower in women than in men (IRwomen: 6.73 vs. IRmen: 14.58 per 1000 person-years). While increase in absolute CVD risk associated with prevalent hypertension was lower in women than in men [WMD: −6.12, 95% confidence interval: (−9.84 to −2.40), P = 0.001], increase in absolute CVD risk associated with prevalent obesity [WMD: −4.25 (−9.11 to 0.61), P = 0.087], type-2 diabetes [WMD: −1.04 (−14.36 to 12.29), P = 0.879] and AF [WMD: 18.39 (−39.65 to 76.43), P = 0.535] did not significantly differ between the sexes. Using relative risk measures, prevalent hypertension [WMR: 1.49%, 95% confidence interval: (1.12–1.99), P = 0.006], type-2 diabetes [WMR: 1.73 (1.09–2.73), P = 0.019], and AF [WMR: 2.53 (1.12–5.70), P = 0.025] were all associated with higher CVD risk in women than in men.ConclusionIncrease in absolute risk of developing CVD is higher in hypertensive men than in hypertensive women, but no substantial sex-related differences were observed among individuals with obesity, type-2 diabetes and AF. On a relative risk scale, comorbidities, in general, confer a higher CVD risk in women than in men.

Obesity, Adiposity, and Risk of Symptomatic Gallstone Disease According to Genetic Susceptibility

Adiposity has been consistently associated with gallstone disease risk. We aimed to characterize associations of anthropometric measures (body mass index [BMI], recent weight change, long-term weight change, waist circumference, and waist-to-hip ratio) with symptomatic gallstone disease according to strata of gallstone disease polygenic risk score (PRS). We conducted analysis among 34,626 participants with available genome-wide genetic data within 3 large, prospective, U.S. cohorts-the Nurses' Health Study (NHS), Health Professionals Follow-Up Study, and NHS II. We characterized joint associations of PRS and anthropometric measures and tested for interactions on the relative and absolute risk scales. Women in the highest BMI and PRS categories (BMI ≥30 kg/m2 and PRS ≥1 SD above mean) had odds ratio for gallstone disease of 5.55 (95% confidence interval, 5.29 to 5.81) compared with those in the lowest BMI and PRS categories (BMI <25 kg/m2 and PRS <1 SD below the mean). The corresponding odds ratio among men was 1.65 (95% confidence interval, 1.02 to 2.29). Associations for BMI did not vary within strata of PRS on the relative risk scale. On the absolute risk scale, the incidence rate difference between obese and normal-weight individuals was 1086 per 100,000 person-years within the highest PRS category, compared with 666 per 100,000 person-years in the lowest PRS category, with strong evidence for interaction with the ABCG8 locus. While maintenance of a healthy body weight reduces gallstone disease risk among all individuals, risk reduction is higher among the subset with greater genetic susceptibility to gallstone disease.

Aortic stiffness and content of adipokines in the serum in persons of European and South Asian ethnic

To evaluate the cardiovascular risk (CVR) based on arterial stiffness and content of adipokines in young-aged persons of different ethnicity (European and South Asian). 290 persons of European (Slavic) and South Asian (Korean) ethnicity aged from 19 to 49 years with and without arterial hypertension (AH) were examined. Clinical, anthropometric, laboratory examinations were performed, levels of resistin and adiponectin of blood were assessed. Total CVR was assessed by SCORE scale, patients under the age of 40 years were assessed by relative risk scale. Aortic stiffness was examined by non-invasive arteriography. Patients of European ethnicity had higher blood pressure (BP), body mass index (BMI), waist circumference (WC), levels of resistin and adiponectin. Pulse wave velocity in the aorta (PWVA) did not differ significantly in ethnic groups. According to the SCORE scale in individuals of the European and South Asian races in general groups and groups with arterial hypertension a moderate absolute risk was determined, in individuals under 40 years of age a moderate relative risk was determined without a significant difference between the groups. However increased levels of PWVA (more than 10 m/s) were registered more often in Korean ethnicity (46.9% compared to Slavic ethnicity, 22.2%). Closer reliable correlations between the level of BP and BMI, WC, PWVA were revealed in Korean ethnicity. Ethnic differences in correlation of adipokines in blood and their dependence on anthropometric and hemodynamic characteristics were described. The assessment of CVR according to traditional scales does not always accurately represent its real level. New information was obtained on the features of adipokine metabolism and its connections with early manifestations of vascular remodeling in young-aged depending on the race. Taking into account ethnic differences, we recommend in-depth diagnostics of CVR in South Asians. The data can be useful for the design of personalized programs for the diagnostics and assessment of CVR.

Rethinking meta-analysis: Addressing problems of non-transportability when combining treatment effects across patient populations

Introduction Standard methodology for meta-analysis (MA) only focuses on deriving a summary estimate but remains implicit about the patient population for which it describes the treatment effect. However, the evidence about treatment effect in one population cannot always be applied to another population that differs possibly in many aspects from that of the original trial, in which case we say there is lack of transportability. In practice, standard MAs do not always ensure homogeneity in the baseline characteristics of the different trial populations; thereby do not rule out a potential lack of transportability. Besides, standard MAs often overlook the under-representation of low- and middle-income countries in the available evidence for several medical interventions. As a consequence, simply meta-analyzing results of trials conducted in developed countries will be not informative for policymakers in low-income countries to implement the most efficient interventions for their populations. Method In this paper, we remedy these issues by proposing an alternative MA approach for randomized clinical trials, which uses individual patient data (IPD) from all trials to infer the treatment effect for the patient population in a given trial, based on case-mix standardization. Denote Y(1k) and Y(0k) as the counterfactual outcome that would be observed for a given individual if (version of) treatment and control used in trial S = k were administered, respectively. Our focus is on estimating the probabilities P{Y(xk) = 1|S = j} (x = 0,1), which are the chance of successfully treated if the patients in population S = j was given the version of treatment (x = 1) or control (x = 0) used in study S = k. Based on these, the effect of the treatment version in population can be measured–for instance, on the relative risk scale RR(j,k) = P{Y(1k) = 1|S = j}/P{Y(0k) = 1|S = j}. Under certain conditions, RR(j, k) can be estimated by two different approaches, namely: – outcome regression; – inverse probability weighting. The estimates log{RR(j, k)} obtained from the same population j are then summarized by conducting a conventional random-effect MA. The resulting summary estimate thereby represents the effect of treatment for the target population j, while the heterogeneity variance estimate expresses how much results from different trials vary even when considered for the same patient population. Results We illustrate the new approach by considering a MA of numerically simulated RCTs that evaluate a binary treatment versus control with respect to a binary outcome. Results of this simulation point out that the proposed framework does enable better understood heterogeneity assessments. The overall heterogeneity across trial results can now be decomposed into 2 different parts, namely case-mix (CM) and beyond case-mix (BCM) heterogeneity. The new framework is then applied to reanalyze a published IPD MA evaluating the effect of vitamin D on the risk of respiratory infection. Finally, we extend the framework to settings where the IPD is only available for some but not for all trials in the MA. Discussion In conclusion, case-mix standardization prior to evidence synthesis is a good strategy to overcome the lack of transportability and other related issues that complicate most MAs. Not only resulting in more interpretable summary results, this strategy also sheds insightful light onto how and why results of different trials are heterogeneous in practice.

Individual and Joint Effects of Pulse Pressure and Blood Pressure Treatment Intensity on Serious Adverse Events in the SPRINT Trial

PurposeThe objective of this study was to determine individual and joint effects of pulse pressure and blood pressure treatment intensity on serious adverse events in the Systolic Blood Pressure Intervention Trial. MethodsPulse pressure was calculated by subtracting diastolic blood pressure from systolic blood pressure. Blood pressure treatment intensity goal was ≤140mm Hg in the control arm and ≤120mm Hg in the intensive arm. The primary outcome was a 5-point composite of hypotension, syncope, electrolyte abnormalities, acute renal insufficiency, or injurious falls. ResultsIn 9361 trial participants, the incident rate for the primary outcome per 1000 person-years increased with higher pulse pressure category: ≤49 mmHg: 20.4 (17.2-24.1), 50-59 mmHg: 24.5 (21.3-28.2), 60-69 mmHg: 31.7 (27.7-36.2), ≥70 mmHg: 44.6 (39.8-49.9; Ptrend < .0001; hazard ratio [HR] of pulse pressure [every 10mm Hg] 1.23; 95% confidence interval [CI], 1.18-1.28). The intensive treatment arm had a higher incidence rate of serious adverse events than the control arm (34.2, 95% CI, 31.2-37.3, vs 26.0, 95% CI, 23.4-28.8, P = .0001; HR 1.32; 95% CI, 1.15-1.51). The combined effect was not significant in the relative risk scale (HR 0.97, Pinteraction = .48) but was significant in the risk difference scale (P = .027), contributing 2.5 additional serious adverse events per 1000 person-years for every 10mm Hg increase in pulse pressure in excess of the individual effects of pulse pressure and treatment intensity. ConclusionsWider pulse pressure and intensive blood pressure treatment were individually associated with the composite adverse event outcome. A modest effect modification of pulse pressure and treatment intensity led to additional adverse events when both were present. Clinicians should use caution when treating older patients with elevated pulse pressure to an intensive blood pressure treatment target.

Case-only approach to identifying markers predicting treatment effects on the relative risk scale.

Retrospectively measuring markers on stored baseline samples from participants in a randomized controlled trial (RCT) may provide high quality evidence as to the value of the markers for treatment selection. Originally developed for approximating gene-environment interactions in the odds ratio scale, the case-only method has recently been advocated for assessing gene-treatment interactions on rare disease endpoints in randomized clinical trials. In this article, the case-only approach is shown to provide a consistent and efficient estimator of marker by treatment interactions and marker-specific treatment effects on the relative risk scale. The prohibitive rare-disease assumption is no longer needed, broadening the utility of the case-only approach. The case-only method is resource-efficient as markers only need to be measured in cases only. It eliminates the need to model the marker's main effect, and can be used with any parametric or nonparametric learning method. The utility of this approach is illustrated by an application to genetic data in the Women's Health Initiative (WHI) hormone therapy trial.

Cumulative occupational shoulder exposures and surgery for subacromial impingement syndrome: a nationwide Danish cohort study

ObjectivesThe primary aim was to examine exposure–response relationships between cumulative occupational shoulder exposures and surgery for subacromial impingement syndrome (SIS), and to compare sex-specific exposure–response relationships. The secondary aim was...

On the “Proportion Eliminated” for Risk Differences Versus Excess Relative Risks

In the January 2013 issue, VanderWeele1 discussed the concept of “proportion eliminated” by fixing intermediates as a policy-relevant proportion for direct effects.2 In the present letter, we discuss the interpretations of and relations between the proportion eliminated for a risk difference and excess relative risk (i.e., relative risk minus 1) and an erratum for the letter of VanderWeele.3 Let X denote a binary exposure of interest, Y a binary outcome, and M a potential mediator. Then, we let Yx(ω) denote the potential outcomes for individual ω if, possibly contrary to fact, there had been interventions to set X to x. We also let Yxm(ω) denote the potential outcomes for individual ω if, possibly contrary to fact, there had been interventions to set X to x and to set M to m. VanderWeele1 defined the proportion eliminated as (TE-CDE(m))/TE, where TE represents the total effect of the exposure on the outcome and CDE(m) represents the controlled direct effect of the exposure on the outcome, intervening to set the intermediate to some fixed level m, where TE and CDE(m) are on the risk different scale. Thus, by using the notations of potential outcomes, the proportion eliminated can be written as {RD(TE) − RD(CDE(m))}/RD(TE) = E[Y1 − Y0) − (Y1m − Y0m)]/E[Y1 − Y0], where RD stands for a risk difference. Note that this measure, although called a “proportion”, is not constrained between 0 and 1. Trivially, this measure is equal to 0 when the risk difference remains identical before and after the intervention on the intermediate (i.e., E[Y1 − Y0] = E[Y1m − Y0m]) whereas it is equal to 1 when one achieves perfect equality between the exposed and unexposed groups by the intervention on the intermediate (i.e., E[Y1m] = E[Y0m]). It is notable that the numerator of this measure can be interpreted as a differential in risk reduction due to the intervention between the exposed and the unexposed groups because it can be rewritten as E[(Y1 − Y1m) − (Y0 − Y0m)]. VanderWeele1 further explained that the “proportion eliminated can also be calculated if a risk ratio scale (or odds ratio scale with a rare outcome) is used to estimate the effects”, providing the following formula {RR(TE)−RR(CDE(m))}/{RR(TE)−1}, where RR stands for a risk ratio. Contrary to the letter of VanderWeele,1 this does not give the PE on the risk difference scale but rather on the excess relative risk scale.3 The two are not equivalent for the proportion eliminated. This can be seen as follows: {RR(TE)−RR(CDE(m))}{RR(TE)−1}={RR(TE)−1}−{RR(CDE(m))−1}{RR(TE)−1}={E[Y1−Y0]/E[Y0]}−{E[Y1m−Y0m]/E[Y0m]}E[Y1−Y0]/E[Y0]≠E[(Y1−Y0)−(Y1m−Y0m)]E[Y1−Y0](∵E[Y0]≠E[Y0m])={RD(TE)−RD(CDE(m))}RD(TE). Thus, the formula yields the PE on the risk difference scale only when E[Y0] = E[Y0m], which will not be the case in general.4,5 One can still interpret the formula as the proportion eliminated on an excess relative risk scale (i.e., by what proportion is the excess relative risk reduced by fixing M to m). Trivially, this measure is equal to 0 when the risk ratio remains identical before and after the intervention on the intermediate (i.e., E[Y1]/E[Y0] = E[Y1m]/E[Y0m]) whereas it is equal to 1 when one achieves perfect equality between the exposed and unexposed groups by the intervention (i.e., E[Y1m] = E[Y0m]). See the eAppendix for some further relations. Finally, it is worth noting that one can still calculate the PE on the risk difference scale from ratio measures.6 The formula that applies when intervening to set a binary intermediate to 0 is: (RERI)P(M=1|X=1)+(RR01−1){P(M=1|X=1)−P(M=1|X=0)}(RR10−1)+(RERI)P(M=1|X=1)+(RR01−1){P(M=1|X=1)−P(M=1|X=0)}, where RRxm = E[Yxm]/E[Y00]) is a risk ratio comparing category X = x, M = m to the reference category X = 0, M = 0, and where RERI or “relative excess risk due to interaction” is a measure of additive interaction using ratios (i.e., RR11 − RR10 − RR01 + 1). See VanderWeele6 for other settings of m and for formulae that are applicable to arbitrary exposures and intermediates. Applying these different formulae of proportions of effect eliminated by fixing intermediates will enhance interpretation of the estimated effect of actual policy interventions.

Introduction of African Swine Fever into the European Union through Illegal Importation of Pork and Pork Products

Transboundary animal diseases can have very severe socio-economic impacts when introduced into new regions. The history of disease incursions into the European Union suggests that initial outbreaks were often initiated by illegal importation of meat and derived products. The European Union would benefit from decision-support tools to evaluate the risk of disease introduction caused by illegal imports in order to inform its surveillance strategy. However, due to the difficulty in quantifying illegal movements of animal products, very few studies of this type have been conducted. Using African swine fever as an example, this work presents a novel risk assessment framework for disease introduction into the European Union through illegal importation of meat and products. It uses a semi-quantitative approach based on factors that likely influence the likelihood of release of contaminated smuggled meat and products, and subsequent exposure of the susceptible population. The results suggest that the European Union is at non-negligible risk of African swine fever introduction through illegal importation of pork and products. On a relative risk scale with six categories from negligible to very high, five European Union countries were estimated at high (France, Germany, Italy and United Kingdom) or moderate (Spain) risk of African swine fever release, five countries were at high risk of exposure if African swine fever were released (France, Italy, Poland, Romania and Spain) and ten countries had a moderate exposure risk (Austria, Bulgaria, Germany, Greece, Hungary, Latvia, Lithuania, Portugal, Sweden and United Kingdom). The approach presented here and results obtained for African swine fever provide a basis for the enhancement of risk-based surveillance systems and disease prevention programmes in the European Union.

Screening-testing approaches for gene-gene and gene-environment interactions using independent statistics

Next-generation sequencing and other high-throughput technologies have made it feasible to characterize millions of sequence variations on large numbers of study participants. But when it comes to identifying a small number of these genetic features (or feature sets) that are associated with a disease trait, the investigator is faced with a formidable multiple-testing challenge. It can be thought of as a signal-tonoise problem, where the large number of unrelated genetic features tends to drown out the faint signal of the small number of biologically relevant features. The theoretical underpinnings of an emerging class of statistical methods for genomic studies, two-stage procedures for both gene-gene and gene-environment interactions have recently been described in a remarkable article (Dai et al., 2012). The key idea is that dimensionality of multiple testing in genomics can be reduced by screening features to be tested with an independent statistic in the same dataset, thereby mitigating the multipletesting problem and increasing power to detect effects. In other words, the noise is reduced, allowing the relevant signal to be more easily detected. These methods will likely gain importance as high-throughput technologies continue to yield exponentially increasing amounts of information per sample and per research dollar spent. Dai et al. couched their paper in the context of gene-environment interactions only. However, it is worth noting that the theoretical properties detailed by Dai et al. apply not just to the search for gene-environment interactions (GxE), but also to (epistatic) interactions between genetic variants (GxG), since in constructing these hypothesis tests, both “gene” and “environment” features are treated analogously as discrete or continuous variables in models designed to identify associations with a disease trait. A notable exception is when the approach depends on the environmental exposure being a randomized treatment, allowingadditional assumptions to be made. One such screening-testing interaction approach is designed for a case-control study where the investigator is interested in identifyingGxGorGxE pairs involved in interactions (Millstein et al., 2006;Murcray et al., 2009; Dai et al., 2012; Lewinger et al., 2013). There is an assumption that each pair of features considered is independent in the general population, and only if a dependence is found in the pooled case-control sample (the screening stage), is the pair tested in a formal model that includes an interaction term (the testing stage), e.g., logit(P[D]) = α+ β1∗SNP1+ β2 ∗SNP2+ β3∗SNP1∗SNP2, where β3 is the interaction parameter and D indicates disease. The interaction parameter can be testedaloneorinamultidegree-of-freedom testofoneorbothmaineffectstogetherwith the interaction, an approach that was generally found to bemore powerful (Millstein et al., 2006; Kraft et al., 2007). An important characteristic of the approach is that even if the independence assumption is not justified, type I error in the testing stage will still be properly controlled. This approach is perhaps more general and more powerful than previously appreciated. The screening procedure appears to be sensitive to both main effects and interactions, not just interactions, as claimed in prior work. The implication is that the approach is less specific to interactions and correspondingly more powerful when main effects are present. In fact, it may be capable of detecting weak interactions coupled with weak main effects. Some authors (Murcray et al., 2009; Dai et al., 2012; Lewinger et al., 2013) have attributed the statistical power of the screening procedure solely to an association in cases due to an interaction in the underlying population (non-zero β3, or more correctly, a departure from multiplicativity on a relative risk scale), as in the case-only interaction analysis (Piegorsch et al., 1994). According to this view, controls only contribute noise to the screening procedure because the factors are independent in this population. Further, if the two features contribute marginal disease risks and a multiplicative relative risk model describes their joint risk, then dependencies will not be induced among cases. The idea is that if there is independence in cases and independence in controls, then it should follow that there would be independence in the pooled case-control sample—but this is not necessarily the case. It has not been adequately appreciated that when cases and controls are pooled, main effects can contribute a substantial increase in power to capture disease-related feature pairs with the above screening procedure. Interestingly, the complex conditioning on disease status inherent in pooling of cases and controls can induce dependencies and thus increase power of the screening procedure when main effects are present. As proof of concept, consider the relatively simple relative risk model, log(P[D]) = λ+ β1∗SNP1+ β2∗SNP2+ β3 ∗SNP1∗SNP2, where exp (λ) is the baseline risk, the two SNPs have equal relative risks per allele, i.e., β1 = β2, there is a weak interaction (small β3), and equal

Translating genomics into improved population screening: hype or hope?

The success of genome wide association studies (GWAS) in identifying common genetic variants that are associated with the risk of common complex diseases has raised expectations about genetic profiling for personalised therapeutic and public health interventions. Questions remain whether genetic profiling can help in targeting screening at genetically defined subpopulations to improve the benefit that harm ratio of screening programmes. Multiple genetic loci contribute to the increased risk of common diseases. For example, to date more than 20 loci have been implicated with type 2 diabetes (McCarthy 2010), more than 30 with prostate cancer and 18 with breast cancer (Varghese and Easton 2010). The risk alleles at these loci individually confer a modest increase in risk of disease (usually per-allele relative risks of less than 1.5) and the risk alleles appear to interact multiplicatively on the relative risk scale (Hingorani et al. 2010). Consequently, the susceptibility variants identified by GWAS explain only a small proportion of heritability of common diseases.

A blood‐based biomarker panel for stratifying current risk for colorectal cancer

Colorectal cancer (CRC) is often curable and preventable using current screening modalities. Unfortunately, screening compliance remains low, partly due to patient dissatisfaction with faecal/endoscopic testing. Recent guidelines advise CRC screening should begin with risk stratification. A blood-based test providing clinically actionable CRC risk information would likely improve screening compliance and enhance clinical decision making. We analyzed 196 gene expression profiles to select candidate CRC biomarkers. qRT-PCR was performed on 642 samples to develop a 7-gene biomarker panel using 112 CRC/120 controls (training set) and 202 CRC/208 controls (independent, blind test set). Panel performance characteristics and disease prevalence (0.7%) were then used to develop a scale assessing an individual's current risk of having CRC based on his/her gene signature. A 7-gene panel (ANXA3, CLEC4D, LMNB1, PRRG4, TNFAIP6, VNN1 and IL2RB) discriminated CRC in the training set (area under the receiver-operating-characteristic curve (ROC AUC), 0.80; accuracy, 73%; sensitivity, 82%; specificity 64%). The independent blind test set confirmed performance (ROC AUC, 0.80; accuracy, 71%; sensitivity, 72%; specificity, 70%). Individual gene profiles were compared against the population results and used to calculate the current relative risk for CRC. We have developed a 7-gene, blood-based biomarker panel that can stratify subjects according to their current relative risk across a broad range in an average-risk population. Across the continuous spectrum of risk as defined by the current relative risk scale, it is possible to identify clinically meaningful reference points that can assist patients and physicians in CRC screening decision making.

The effect of joint exposures: examining the presence of interaction

Clinical epidemiological studies investigate whether an exposure, or risk factor, is causally related to the development or progression of a disease or mortality. It might be of interest to study whether this relation is different in different types of patients. To address such research questions, the presence of interaction among risk factors can be examined. Causal interaction between two risk factors is considered most clinically relevant in epidemiology. Causal interaction occurs when two risk factors act together in causing disease and is explicitly defined as a deviation from additivity on a risk difference scale. Statistical interaction can be evaluated on both an additive (absolute risk) and multiplicative (relative risk) scale, depending on the model that is used. When using logistic regression models, which are multiplicative models, several measures of additive interaction are presented to evaluate whether the magnitude of an association differs across subgroups: the relative excess risk due to interaction (RERI), the attributable proportion due to interaction (AP), or the synergy index (S). For a transparent presentation of interaction effects the recent Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement advises reporting the separate effect of each exposure as well as the joint effect compared with the unexposed group as a joint reference category to permit evaluation of both additive and multiplicative interaction.