How to use machine learning and fuzzy cognitive maps to test hypothetical scenarios in health behavior change interventions: a case study on fruit intake.

BACKGROUND: Intervention planners use logic models to design evidence-based health behavior interventions. Logic models that capture the complexity of health behavior necessitate additional computational techniques to inform decisions with respect to the design of interventions. OBJECTIVE: Using empirical data from a real intervention, the present paper demonstrates how machine learning can be used together with fuzzy cognitive maps to assist in designing health behavior change interventions. METHODS: A modified Real Coded Genetic algorithm was applied on longitudinal data from a real intervention study. The dataset contained information about 15 determinants of fruit intake among 257 adults in the Netherlands. Fuzzy cognitive maps were used to analyze the effect of two hypothetical intervention scenarios designed by domain experts. RESULTS: Simulations showed that the specified hypothetical interventions would have small impact on fruit intake. The results are consistent with the empirical evidence used in this paper. CONCLUSIONS: Machine learning together with fuzzy cognitive maps can assist in building health behavior interventions with complex logic models. The testing of hypothetical scenarios may help interventionists finetune the intervention components thus increasing their potential effectiveness.

A systems science approach to identifying data gaps in national data sources on adolescent suicidal ideation and suicide attempt in the United States.

BACKGROUND: Suicide is currently the second leading cause of death among adolescents ages 10-14, and third leading cause of death among adolescents ages 15-19 in the United States (U.S). Although we have numerous U.S. based surveillance systems and survey data sources, the coverage offered by these data with regard to the complexity of youth suicide had yet to be examined. The recent release of a comprehensive systems map for adolescent suicide provides an opportunity to contrast the content of surveillance systems and surveys with the mechanisms listed in the map. OBJECTIVE: To inform existing data collection efforts and advance future research on the risk and protective factors relevant to adolescent suicide. METHODS: We examined data from U.S. based surveillance systems and nationally-representative surveys that included (1) observations for an adolescent population and (2) questions or indicators in the data that identified suicidal ideation or suicide attempt. Using thematic analysis, we evaluated the codebooks and data dictionaries for each source to match questions or indicators to suicide-related risk and protective factors identified through a recently published suicide systems map. We used descriptive analysis to summarize where data were available or missing and categorized data gaps by social-ecological level. RESULTS: Approximately 1-of-5 of the suicide-related risk and protective factors identified in the systems map had no supporting data, in any of the considered data sources. All sources cover less than half the factors, except the Adolescent Brain Cognitive Development Study (ABCD), which covers nearly 70% of factors. CONCLUSIONS: Examining gaps in suicide research can help focus future data collection efforts in suicide prevention. Our analysis precisely identified where data is missing and also revealed that missing data affects some aspects of suicide research (e.g., distal factors at the community and societal level) more than others (e.g., proximal factors about individual characteristics). In sum, our analysis highlights limitations in current suicide-related data availability and provides new opportunities to identify and expand current data collection efforts.

Identifying small groups of foods that can predict achievement of key dietary recommendations: data mining of the UK National Diet and Nutrition Survey, 2008-12.

OBJECTIVE: Many dietary assessment methods attempt to estimate total food and nutrient intake. If the intention is simply to determine whether participants achieve dietary recommendations, this leads to much redundant data. We used data mining techniques to explore the number of foods that intake information was required on to accurately predict achievement, or not, of key dietary recommendations. DESIGN: We built decision trees for achievement of recommendations for fruit and vegetables, sodium, fat, saturated fat and free sugars using data from a national dietary surveillance data set. Decision trees describe complex relationships between potential predictor variables (age, sex and all foods listed in the database) and outcome variables (achievement of each of the recommendations). SETTING: UK National Diet and Nutrition Survey (NDNS, 2008-12). SUBJECTS: The analysis included 4156 individuals. RESULTS: Information on consumption of 113 out of 3911 (3 %) foods, plus age and sex was required to accurately categorize individuals according to all five recommendations. The best trade-off between decision tree accuracy and number of foods included occurred at between eleven (for fruit and vegetables) and thirty-two (for fat, plus age) foods, achieving an accuracy of 72 % (for fat) to 83 % (for fruit and vegetables), with similar values for sensitivity and specificity. CONCLUSIONS: Using information on intake of 113 foods, it is possible to predict with 72-83 % accuracy whether individuals achieve key dietary recommendations. Substantial further research is required to make use of these findings for dietary assessment.

Identifying binge drinkers based on parenting dimensions and alcohol-specific parenting practices: building classifiers on adolescent-parent paired data.

BACKGROUND: Most Dutch adolescents aged 16 to 18 engage in binge drinking. Previous studies have investigated how parenting dimensions and alcohol-specific parenting practices are related to adolescent alcohol consumption. Mixed results have been obtained on both dimensions and practices, highlighting the complexity of untangling alcohol-related factors. The aim of this study was to investigate (1) whether parents' reports of parenting dimensions and alcohol-specific parenting practices, adolescents' perceptions of these dimensions and practices, or a combination are most informative to identify binge drinkers, and (2) which of these parenting dimensions and alcohol-specific parenting practices are most informative to identify binge drinkers. METHODS: Survey data of 499 adolescent-parent dyads were collected. The computational technique of data mining was used to allow for a data driven exploration of nonlinear relationships. Specifically, a binary classification task, using an alternating decision tree, was conducted and measures regarding the performance of the classifiers are reported after a 10-fold cross-validation. RESULTS: Depending on the parenting dimension or practice, parents' reports correctly identified the drinking behaviour of 55.8% (using psychological control) up to 70.2% (using rules) of adolescents. Adolescents' perceptions were best at identifying binge drinkers whereas parents' perceptions were best at identifying non-binge drinkers. CONCLUSIONS: Of the parenting dimensions and practices, rules are particularly informative in understanding drinking behaviour. Adolescents' perceptions and parents' reports are complementary as they can help identifying binge drinkers and non-binge drinkers respectively, indicating that surveying specific aspects of adolescent-parent dynamics can improve our understanding of complex addictive behaviours.

Impact of different policies on unhealthy dietary behaviors in an urban adult population: an agent-based simulation model.

OBJECTIVES: Unhealthy eating is a complex-system problem. We used agent-based modeling to examine the effects of different policies on unhealthy eating behaviors. METHODS: We developed an agent-based simulation model to represent a synthetic population of adults in Pasadena, CA, and how they make dietary decisions. Data from the 2007 Food Attitudes and Behaviors Survey and other empirical studies were used to calibrate the parameters of the model. Simulations were performed to contrast the potential effects of various policies on the evolution of dietary decisions. RESULTS: Our model showed that a 20% increase in taxes on fast foods would lower the probability of fast-food consumption by 3 percentage points, whereas improving the visibility of positive social norms by 10%, either through community-based or mass-media campaigns, could improve the consumption of fruits and vegetables by 7 percentage points and lower fast-food consumption by 6 percentage points. Zoning policies had no significant impact. CONCLUSIONS: Interventions emphasizing healthy eating norms may be more effective than directly targeting food prices or regulating local food outlets. Agent-based modeling may be a useful tool for testing the population-level effects of various policies within complex systems.

Creating groups with similar expected behavioural response in randomized controlled trials: a fuzzy cognitive map approach.

BACKGROUND: Controlling bias is key to successful randomized controlled trials for behaviour change. Bias can be generated at multiple points during a study, for example, when participants are allocated to different groups. Several methods of allocations exist to randomly distribute participants over the groups such that their prognostic factors (e.g., socio-demographic variables) are similar, in an effort to keep participants' outcomes comparable at baseline. Since it is challenging to create such groups when all prognostic factors are taken together, these factors are often balanced in isolation or only the ones deemed most relevant are balanced. However, the complex interactions among prognostic factors may lead to a poor estimate of behaviour, causing unbalanced groups at baseline, which may introduce accidental bias. METHODS: We present a novel computational approach for allocating participants to different groups. Our approach automatically uses participants' experiences to model (the interactions among) their prognostic factors and infer how their behaviour is expected to change under a given intervention. Participants are then allocated based on their inferred behaviour rather than on selected prognostic factors. RESULTS: In order to assess the potential of our approach, we collected two datasets regarding the behaviour of participants (n = 430 and n = 187). The potential of the approach on larger sample sizes was examined using synthetic data. All three datasets highlighted that our approach could lead to groups with similar expected behavioural changes. CONCLUSIONS: The computational approach proposed here can complement existing statistical approaches when behaviours involve numerous complex relationships, and quantitative data is not readily available to model these relationships. The software implementing our approach and commonly used alternatives is provided at no charge to assist practitioners in the design of their own studies and to compare participants' allocations.

Simulation Models for Suicide Prevention: A Survey of the State-of-the-Art.

Suicide is a leading cause of death and a global public health problem, representing more than one in every 100 deaths in 2019. Modeling and Simulation (M&S) is widely used to address public health problems, and numerous simulation models have investigated the complex, dependent, and dynamic risk factors contributing to suicide. However, no review has been dedicated to these models, which prevents modelers from effectively learning from each other and raises the risk of redundant efforts. To guide the development of future models, in this paper we perform the first scoping review of simulation models for suicide prevention. Examining ten articles, we focus on three practical questions. First, which interventions are supported by previous models? We found that four groups of models collectively support 53 interventions. We examined these interventions through the lens of global recommendations for suicide prevention, highlighting future areas for model development. Second, what are the obstacles preventing model application? We noted the absence of cost effectiveness in all models reviewed, meaning that certain simulated interventions may be infeasible. Moreover, we found that most models do not account for different effects of suicide prevention interventions across demographic groups. Third, how much confidence can we place in the models? We evaluated models according to four best practices for simulation, leading to nuanced findings that, despite their current limitations, the current simulation models are powerful tools for understanding the complexity of suicide and evaluating suicide prevention interventions.

When Do We Need Massive Computations to Perform Detailed COVID-19 Simulations?

The COVID-19 pandemic has infected over 250 million people worldwide and killed more than 5 million as of November 2021. Many intervention strategies are utilized (e.g., masks, social distancing, vaccinations), but officials making decisions have a limited time to act. Computer simulations can aid them by predicting future disease outcomes, but they also require significant processing power or time. It is examined whether a machine learning model can be trained on a small subset of simulation runs to inexpensively predict future disease trajectories resembling the original simulation results. Using four previously published agent-based models (ABMs) for COVID-19, a decision tree regression for each ABM is built and its predictions are compared to the corresponding ABM. Accurate machine learning meta-models are generated from ABMs without strong interventions (e.g., vaccines, lockdowns) using small amounts of simulation data: the root-mean-square error (RMSE) with 25% of the data is close to the RMSE for the full dataset (0.15 vs 0.14 in one model; 0.07 vs 0.06 in another). However, meta-models for ABMs employing strong interventions require much more training data (at least 60%) to achieve a similar accuracy. In conclusion, machine learning meta-models can be used in some scenarios to assist in faster decision-making.

An Experimental Study on the Scalability of Recent Node Centrality Metrics in Sparse Complex Networks.

Node centrality measures are among the most commonly used analytical techniques for networks. They have long helped analysts to identify "important" nodes that hold power in a social context, where damages could have dire consequences for transportation applications, or who should be a focus for prevention in epidemiology. Given the ubiquity of network data, new measures have been proposed, occasionally motivated by emerging applications or by the ability to interpolate existing measures. Before analysts use these measures and interpret results, the fundamental question is: are these measures likely to complete within the time window allotted to the analysis? In this paper, we comprehensively examine how the time necessary to run 18 new measures (introduced from 2005 to 2020) scales as a function of the number of nodes in the network. Our focus is on giving analysts a simple and practical estimate for sparse networks. As the time consumption depends on the properties in the network, we nuance our analysis by considering whether the network is scale-free, small-world, or random. Our results identify that several metrics run in the order of O(nlogn) and could scale to large networks, whereas others can require O(n 2) or O(n 3) and may become prime targets in future works for approximation algorithms or distributed implementations.

A Scoping Review of Three Dimensions for Long-Term COVID-19 Vaccination Models: Hybrid Immunity, Individual Drivers of Vaccinal Choice, and Human Errors.

The virus that causes COVID-19 changes over time, occasionally leading to Variants of Interest (VOIs) and Variants of Concern (VOCs) that can behave differently with respect to detection kits, treatments, or vaccines. For instance, two vaccination doses were 61% effective against the BA.1 predominant variant, but only 24% effective when BA.2 became predominant. While doses still confer protection against severe disease outcomes, the BA.5 variant demonstrates the possibility that individuals who have received a few doses built for previous variants can still be infected with newer variants. As previous vaccines become less effective, new ones will be released to target specific variants and the whole process of vaccinating the population will restart. While previous models have detailed logistical aspects and disease progression, there are three additional key elements to model COVID-19 vaccination coverage in the long term. First, the willingness of the population to participate in regular vaccination campaigns is essential for long-term effective COVID-19 vaccination coverage. Previous research has shown that several categories of variables drive vaccination status: sociodemographic, health-related, psychological, and information-related constructs. However, the inclusion of these categories in future models raises questions about the identification of specific factors (e.g., which sociodemographic aspects?) and their operationalization (e.g., how to initialize agents with a plausible combination of factors?). While previous models separately accounted for natural- and vaccine-induced immunity, the reality is that a significant fraction of individuals will be both vaccinated and infected over the coming years. Modeling the decay in immunity with respect to new VOCs will thus need to account for hybrid immunity. Finally, models rarely assume that individuals make mistakes, even though this over-reliance on perfectly rational individuals can miss essential dynamics. Using the U.S. as a guiding example, our scoping review summarizes these aspects (vaccinal choice, immunity, and errors) through ten recommendations to support the modeling community in developing long-term COVID-19 vaccination models.