What is the most appropriate follow-up time for detecting the epidemiological relationship between coronary artery disease and its main risk factors: novel findings from a 35-year follow-up study.

BACKGROUND: The aim was to investigate the most appropriate follow-up time to detect the associations of coronary artery disease (CAD) with its traditional risk factors in a long-term prospective cohort study. METHODS: The Kuopio Ischaemic Heart Disease Risk Factors Study provided the study material of 1958 middle-aged men free from CAD at baseline and followed up for 35 years. We performed Cox models adjusted for age, family history, diabetes, obesity, hypercholesterolemia, hypertension, smoking, and physical activity, investigated covariate interactions, and tested Schoenfeld residuals to detect time-dependent covariates. Moreover, we applied a sliding window procedure with a subarray of 5 years to better differentiate between risk factors manifested within years and those manifested within decades. The investigated manifestations were CAD and fatal acute myocardial infarction (AMI). RESULTS: Seven hundred seventeen (36.6%) men had CAD, and 109 (5.6%) men died from AMI. After 10 years of follow-up, diabetes became the strongest predictor of CAD with a fully adjusted hazard ratio (HR) of 2.5-2.8. During the first 5 years, smoking was the strongest predictor (HR 3.0-3.8). When the follow-up time was 8-19 years, hypercholesterolemia predicted CAD with a HR of >2. The associations of CAD with age and diabetes depended on time. Age hypertension was the only statistically significant covariate interaction. The sliding window procedure highlighted the significance of diabetes over the first 20 years and hypertension after that. Regarding AMI, smoking was associated with the highest fully adjusted HR (2.9-10.1) during the first 13 years. The associations of extreme and low physical activity with AMI peaked when the follow-up time was 3-8 years. Diabetes showed its highest HR (2.7-3.7) when the follow-up time was 10-20 years. During the last 16 years, hypertension was the strongest predictor of AMI (HR 3.1-6.4). CONCLUSION: The most appropriate follow-up time for most CAD risk factors was 10-20 years. Concerning smoking and hypertension shorter and longer follow-up times could be considered, respectively, particularly when studying fatal AMI. In general, prospective cohort studies of CAD would provide more comprehensive results by reporting point estimates in relation to more than one timepoint and concerning sliding windows.

Epidemiological analysis of coronary heart disease and its main risk factors: are their associations multiplicative, additive, or interactive?

OBJECTIVE: The purpose of this study was to discover how considering multiplicative, additive, and interactive effects modifies results of a prospective cohort study on coronary heart disease (CHD) incidence and its main risk factors. MATERIAL AND METHODS: The Kuopio Ischaemic Heart Disease Risk Factor (KIHD) Study provided the study material, 2682 Eastern Finnish middle-aged men, followed since the 1980s. We applied multiplicative and additive survival models together with different statistical metrics and confidence intervals for risk ratios and risk differences to estimate the nature of associations. RESULTS: The mean (SD) follow-up time among men who were free of CHD at baseline (n = 1958) was 21.4 (10.4) years, and 717 (37%) of them had the disease and 301 (15%) died for CHD before the end of follow-up. All tested non-modifiable and modifiable risk factors statistically significantly predicted CHD incidence. We detected three interactions: circulating low-density lipoprotein cholesterol (LDL-C) × age, obesity × age, and obesity × smoking of which LDL-C × age was the most evident one. High LDL-C increased the risk of CHD more among men younger than 50 [risk ratio (RR) 2.10] than those older than 50 (RR 1.22). LDL-C status was the only additive covariate. The additive effect of high LDL-C increased almost linearly up to 18 years and then reached a plateau. The simple multiplicative survival model stressed glycemic status as the strongest modifiable risk factor for developing CHD [hazard ratio (HR) for diabetes vs. normoglycemia was 2.69], whereas the model considering interactions and time dependence emphasised the role of LDL-C status (HR for high LDL-C vs. lower than borderline was 4.43). Age was the strongest non-modifiable predictor. CONCLUSIONS: Including covariate interactions and time dependence in survival models potentially refine results of epidemiological analyses and ease to define the order of importance across CHD risk factors. KEY MESSAGESIncluding covariate interactions and time dependence in survival models potentially refine results of epidemiological analyses on coronary heart disease.Including covariate interactions and time dependence in survival models potentially ease to define the order of importance across coronary heart disease risk factors.

Epidemiological predictive modeling: lessons learned from the Kuopio ischemic heart disease risk factor study.

PURPOSE: The use of predictive models in epidemiology is relatively narrow as most of the studies report results of traditional statistical models such as Linear, Logistic, or Cox regressions. In this study, a high-dimensional epidemiological cohort, collected within the Kuopio Ischemic Heart Disease Risk Factor Study in 1984-1989, was used to investigate the predictive ability of models with embedded variable selection. METHODS: Simple Logistic Regression with seven preselected risk factors was compared to k-Nearest Neighbors, Logistic Lasso Regression, Decision Tree, Random Forest, and Multilayer Perceptron in predicting cardiovascular death for the aged men from Kuopio Ischemic Heart Disease Risk Factor for the long horizon of 30 ± 3 years: 746 predictor variables were available for 2682 men (705 cardiovascular deaths were registered). We considered two scenarios of handling competing risks (removing subjects and treating them as non-cases). RESULTS: The best average AUC on the test sample was 0.8075 (95%CI, 0.8051-0.8099) in scenario 1 and 0.7155 (95%CI, 0.7128-0.7183) in scenario 2 achieved with Logistic Lasso Regression, which was 6.04% and 5.50% higher than the baseline AUC provided by Logistic Regression with manually preselected predictors. CONCLUSIONS: In both scenarios Logistic Lasso Regression, Random Forest, and Multilayer Perceptron outperformed Simple Logistic Regression.

Post-Analysis of Predictive Modeling with an Epidemiological Example.

Post-analysis of predictive models fosters their application in practice, as domain experts want to understand the logic behind them. In epidemiology, methods explaining sophisticated models facilitate the usage of up-to-date tools, especially in the high-dimensional predictor space. Investigating how model performance varies for subjects with different conditions is one of the important parts of post-analysis. This paper presents a model-independent approach for post-analysis, aiming to reveal those subjects' conditions that lead to low or high model performance, compared to the average level on the whole sample. Conditions of interest are presented in the form of rules generated by a multi-objective evolutionary algorithm (MOGA). In this study, Lasso logistic regression (LLR) was trained to predict cardiovascular death by 2016 using the data from the 1984-1989 examination within the Kuopio Ischemic Heart Disease Risk Factor Study (KIHD), which contained 2682 subjects and 950 preselected predictors. After 50 independent runs of five-fold cross-validation, the model performance collected for each subject was used to generate rules describing "easy" and "difficult" cases. LLR with 61 selected predictors, on average, achieved 72.53% accuracy on the whole sample. However, during post-analysis, three categories of subjects were discovered: "Easy" cases with an LLR accuracy of 95.84%, "difficult" cases with an LLR accuracy of 48.11%, and the remaining cases with an LLR accuracy of 71.00%. Moreover, the rule analysis showed that medication was one of the main confusing factors that led to lower model performance. The proposed approach provides insightful information about subjects' conditions that complicate predictive modeling.

Effects of data preprocessing on results of the epidemiological analysis of coronary heart disease and behaviour-related risk factors.

BACKGROUND: We carried out this study to demonstrate the effects of outcome sensitivity, participant exclusions, and covariate manipulations on results of the epidemiological analysis of coronary heart disease (CHD) and its behaviour-related risk factors. MATERIAL AND METHODS: Our study population consisted of 1592 54-year-old men, who participated in the Kuopio Ischaemic Heart Disease Risk Factor (KIHD) Study. We used the Cox proportional-hazards model to predict the hazard of CHD and applied different sets of outcomes concerning outcome sensitivity and data preprocessing procedures regarding participant exclusions and covariate manipulations. RESULTS: The mean follow-up time was 23 years, and 730 men received the CHD diagnosis. Cox regressions based on data with no participant exclusions most often discovered statistically significant associations. Loose inclusion criteria for study participants with any CVD during the follow-up and strict exclusion criteria for participants with no CVD were best in discovering the associations between risk factors and CHD. Outcome sensitivity affected the associations, whereas the covariate type, continuous or categorical, did not. CONCLUSIONS: This study suggests that excluding study participants who are not disease-free at baseline is probably unnecessary for epidemiological analyses. Epidemiological research reports should present results based on no data exclusions together with results based on reasoned exclusions.

Potential and limitations of a pilot-scale drinking water distribution system for bacterial community predictive modelling.

Waterborne disease outbreaks are a persistent and serious threat to public health according to reported incidents across the globe. Online drinking water quality monitoring technologies have evolved substantially and have become more accurate and accessible. However, using online measurements alone is unsuitable for detecting microbial regrowth, potentially including harmful species, ahead of time in the distribution systems. Alternatively, observational data could be collected periodically, e.g. once per week or once per month and it could include a representative set of variables: physicochemical water characteristics, disinfectant concentrations, and bacterial abundances, which would be a valuable source of knowledge for predictive modelling that aims to reveal pathogen-related threats. In this study, we utilised data collected from a pilot-scale drinking water distribution system. A data-driven random forest model was used for predictive modelling and was trained for nowcasting and forecasting abundances of bacterial groups. In all the experiments, we followed the realistic crossline scenario, which means that when training and testing the models the data is collected from different pipelines. In spite of the more accurate results of the nowcasting, the 1-week forecasting still provided accurate predictions of the most abundant bacteria, their rapid increase and decrease. In the future predictive modelling might be used as a tool in designing control measures for opportunistic pathogens which are able to multiply in the favourable conditions in drinking water distribution systems (DWDS). Eventually, the forecasting information will be able to produce practically helpful data for controlling the DWDS regrowth.

Evolutionary methods for variable selection in the epidemiological modeling of cardiovascular diseases.

BACKGROUND: The redundancy of information is becoming a critical issue for epidemiologists. High-dimensional datasets require new effective variable selection methods to be developed. This study implements an advanced evolutionary variable selection method which is applied for cardiovascular predictive modeling. The epidemiological follow-up study KIHD (Kuopio Ischemic Heart Disease Risk Factor Study) was used to compare the designed variable selection method based on an evolutionary search with conventional stepwise selection. The sample contains in total 433 predictor variables and a response variable indicating incidents of cardiovascular diseases for 1465 study subjects. RESULTS: The effectiveness of variable selection methods was investigated in combination with two models: Generalized Linear Logistic Regression and Support Vector Machine. We managed to decrease the number of variables from 433 to 38 and save the predictive ability of the models used. Their performance was evaluated with an F-score metric. At most, we gained 65.6% and 67.4% of the F-score before and after variable selection respectively. All the results were averaged over 5-folds of a cross-validation procedure. CONCLUSIONS: The presented evolutionary variable selection method allows a reduced set of variables to be chosen which are relevant to predicting cardiovascular diseases. A reference list of the most meaningful variables is introduced to be used as a basis for new epidemiological studies. In general, the multicollinearity of variables enables different combinations of predictors to be used and the same performance of models to be attained.