A narrative review of the use of agent-based modeling in health behavior and behavior intervention

The iterative development and refinement of health psychology theories through formal, dynamical systems modelling: a scoping review and initial expert-derived 'best practice' recommendations

This scoping review aimed to synthesise methodological steps taken by researchers in the development of formal, dynamical systems models of health psychology theories. We searched MEDLINE, PsycINFO, the ACM Digital Library and IEEE Xplore in July 2023. We included studies of any design providing that they reported on the development or refinement of a formal, dynamical systems model unfolding at the within-person level, with no restrictions on population or setting. A narrative synthesis with frequency analyses was conducted. A total of 17 modelling projects reported across 29 studies were included. Formal modelling efforts have largely been concentrated to a small number of interdisciplinary teams in the United States (79.3%). The models aimed to better understand dynamic processes (69.0%) or inform the development of adaptive interventions (31.0%). Models typically aimed to formalise the Social Cognitive Theory (31.0%) or the Self-Regulation Theory (17.2%) and varied in complexity (range: 3-30 model components). Only 3.4% of studies reported involving stakeholders in the modelling process and 10.3% drew on Open Science practices. We conclude by proposing an initial set of expert-derived 'best practice' recommendations. Formal, dynamical systems modelling is poised to help health psychologists develop and refine theories, ultimately leading to more potent interventions.

Modelling HIV/AIDS epidemiological complexity: A scoping review of Agent-Based Models and their application

OBJECTIVE

To end the AIDS epidemic by 2030, despite the increasing poverty and inequalities, policies should be designed to deal with population heterogeneity and environmental changes. Bottom-up designs, such as the Agent-Based Model (ABM), can model these features, dealing with such complexity. HIV/AIDS has a complex dynamic of structural factors, risk behaviors, biomedical characteristics and interventions. All embedded in unequal, stigmatized and heterogeneous social structure. To understand how ABMs can model this complexity, we performed a scoping review of HIV applications, highlighting their potentialities.

METHODS

We searched on PubMed, Web of Science, and Scopus repositories following the PRISMA extension for scoping reviews. Our inclusion criteria were HIV/AIDS studies with an ABM application. We identified the main articles using a local co-citation analysis and categorized the overall literature aims, (sub)populations, regions, and if the papers declared the use of ODD protocol and limitations.

RESULTS

We found 154 articles. We identified eleven main papers, and discussed them using the overall category results. Most studies model Transmission Dynamics (37/154), about Men who have sex with Men (MSM) (41/154), or individuals living in the US or South Africa (84/154). Recent studies applied ABM to model PrEP interventions (17/154) and Racial Disparities (12/154). Only six papers declared the use of ODD Protocol (6/154), and 34/154 didn't mention the study limitations.

CONCLUSIONS

While ABM is among the most sophisticated techniques available to model HIV/AIDS complexity. Their applications are still restricted to some realities. However, researchers are challenged to think about social structure due model characteristics, the inclusion of these features is still restricted to case-specific. Data and computational power availability can enhance this feature providing insightful results.

The long-term effectiveness and cost-effectiveness of public health interventions; how can we model behavior? A review

The effectiveness and cost of a public health intervention is dependent on complex human behaviors, yet health economic models typically make simplified assumptions about behavior, based on little theory or evidence. This paper reviews existing methods across disciplines for incorporating behavior within simulation models, to explore what methods could be used within health economic models and to highlight areas for further research. This may lead to better-informed model predictions. The most promising methods identified which could be used to improve modeling of the causal pathways of behavior-change interventions include econometric analyses, structural equation models, data mining and agent-based modeling; the latter of which has the advantage of being able to incorporate the non-linear, dynamic influences on behavior, including social and spatial networks. Twenty-two studies were identified which quantify behavioral theories within simulation models. These studies highlight the importance of combining individual decision making and interactions with the environment and demonstrate the importance of social norms in determining behavior. However, there are many theoretical and practical limitations of quantifying behavioral theory. Further research is needed about the use of agent-based models for health economic modeling, and the potential use of behavior maintenance theories and data mining.

Opioid-related harms and care impacts of conventional and AI-based prescription management strategies: insights from leveraging agent-based modeling and machine learning

Introduction

Like its counterpart to the south, Canada ranks among the top five countries with the highest rates of opioid prescriptions. With many suffering from opioid use disorder first having encountered opioids via prescription routes, practitioners and health systems have an enduring need to identify and effectively respond to the problematic use of opioid prescription. There are strong challenges to successfully addressing this need: importantly, the patterns of prescription fulfillment that signal opioid abuse can be subtle and difficult to recognize, and overzealous enforcement can deprive those with legitimate pain management needs the appropriate care. Moreover, injudicious responses risk shifting those suffering from early-stage abuse of prescribed opioids to illicitly sourced street alternatives, whose varying dosage, availability, and the risk of adulteration can pose grave health risks.

Methods

This study employs a dynamic modeling and simulation to evaluate the effectiveness of prescription regimes employing machine learning monitoring programs to identify the patients who are at risk of opioid abuse while being treated with prescribed opioids. To this end, an agent-based model was developed and implemented to examine the effect of reduced prescribing and prescription drug monitoring programs on overdose and escalation to street opioids among patients, and on the legitimacy of fulfillments of opioid prescriptions over a 5-year time horizon. A study released by the Canadian Institute for Health Information was used to estimate the parameter values and assist in the validation of the existing agent-based model.

Results and discussion

The model estimates that lowering the prescription doses exerted the most favorable impact on the outcomes of interest over 5 years with a minimum burden on patients with a legitimate need for pharmaceutical opioids. The accurate conclusion about the impact of public health interventions requires a comprehensive set of outcomes to test their multi-dimensional effects, as utilized in this research. Finally, combining machine learning and agent-based modeling can provide significant advantages, particularly when using the latter to gain insights into the long-term effects and dynamic circumstances of the former.

A Systematic Review of Systems Science Approaches to Understand and Address Domestic and Gender-Based Violence

Purpose

We aimed to synthesize insights from systems science approaches applied to domestic and gender-based violence.

Methods

We conducted a systematic review of systems science studies (systems thinking, group model-building, agent-based modeling [ABM], system dynamics [SD] modeling, social network analysis [SNA], and network analysis [NA]) applied to domestic or gender-based violence, including victimization, perpetration, prevention, and community responses. We used blinded review to identify papers meeting our inclusion criteria (i.e., peer-reviewed journal article or published book chapter that described a systems science approach to domestic or gender-based violence, broadly defined) and assessed the quality and transparency of each study.

Results

Our search yielded 1,841 studies, and 74 studies met our inclusion criteria (45 SNA, 12 NA, 8 ABM, and 3 SD). Although research aims varied across study types, the included studies highlighted social network influences on risks for domestic violence, clustering of risk factors and violence experiences, and potential targets for intervention. We assessed the quality of the included studies as moderate, though only a minority adhered to best practices in model development and dissemination, including stakeholder engagement and sharing of model code.

Conclusions

Systems science approaches for the study of domestic and gender-based violence have shed light on the complex processes that characterize domestic violence and its broader context. Future research in this area should include greater dialogue between different types of systems science approaches, consideration of peer and family influences in the same models, and expanded use of best practices, including continued engagement of community stakeholders.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10896-023-00578-8.

Using systems science to advance health equity in tobacco control: a causal loop diagram of smoking

OBJECTIVES

Develop and use a causal loop diagram (CLD) of smoking among racial/ethnic minority and lower-income groups to anticipate the intended and unintended effects of tobacco control policies.

METHODS

We developed a CLD to elucidate connections between individual, environmental and structural causes of racial/ethnic and socioeconomic disparities in smoking. The CLD was informed by a review of conceptual and empirical models of smoking, fundamental cause and social stress theories and 19 qualitative interviews with tobacco control stakeholders. The CLD was then used to examine the potential impacts of three tobacco control policies.

RESULTS

The CLD includes 24 constructs encompassing individual (eg, risk perceptions), environmental (eg, marketing) and structural (eg, systemic racism) factors associated with smoking. Evaluations of tobacco control policies using the CLD identified potential unintended consequences that may maintain smoking disparities. For example, the intent of a smoke-free policy for public housing is to reduce smoking among residents. Our CLD suggests that the policy may reduce smoking among residents by reducing smoking among family/friends, which subsequently reduces pro-smoking norms and perceptions of tobacco use as low risk. On the other hand, some residents who smoke may violate the policy. Policy violations may result in financial strain and/or housing instability, which increases stress and reduces feelings of control, thus having the unintended consequence of increasing smoking.

CONCLUSIONS

The CLD may be used to support stakeholder engagement in action planning and to identify non-traditional partners and approaches for tobacco control.

A Multi-Pronged Evaluation of a Healthy Food Access Initiative in Central Texas: Study Design, Methods, and Baseline Findings of the FRESH-Austin Evaluation Study

Food insecurity and limited healthy food access are complex public health issues and warrant multi-level evaluations. The purpose of this paper was to present the overall study design and baseline results of the multi-pronged evaluation of a healthy food access (i.e., Fresh for Less (FFL)) initiative in Central Texas. The 2018-2021 FRESH-Austin study was a natural experiment that utilized a cluster random sampling strategy to recruit three groups of participants (total n = 400): (1) customers at FFL assets, (2) residents that lived within 1.5 miles of an FFL asset, and (3) residents from a comparison community. Evaluation measures included annual cohort surveys, accelerometers and GPS devices, store-level audits, and built environment assessments. Data are being used to inform and validate an agent-based model (ABM) to predict food shopping and consumption behaviors. Sociodemographic factors and food shopping and consumption behaviors were similar across the three groups; however, customers recruited at FFL assets were lower income and had a higher prevalence of food insecurity. The baseline findings demonstrate the need for multi-level food access interventions, such as FFL, in low-income communities. In the future, ABM can be used as a cost-effective way to determine potential impacts of future large-scale food environment programs and policies.

Theory-Guided Randomized Neural Networks for Decoding Medication-Taking Behavior

Long-term endocrine therapy (e.g. Tamoxifen, aromatase inhibitors) is crucial to prevent breast cancer recurrence, yet rates of adherence to these medications are low. To develop, evaluate, and sustain future interventions, individual-level modeling can be used to understand breast cancer survivors' behavioral mechanisms of medication-taking. This paper presents interdisciplinary research, wherein a model employing randomized neural networks was developed to predict breast cancer survivors' daily medication-taking behavior based on their survey data over three time periods (baseline, 4 months, 8 months). The neural network structure was guided by random utility theory developed in psychology and behavioral economics. Comparative analysis indicates that the proposed model outperforms existing computational models in terms of prediction accuracy under conditions of randomness.

The dynamics of food shopping behavior: Exploring travel patterns in low-income Detroit neighborhoods experiencing extreme disinvestment using agent-based modeling

Only a handful of studies have leveraged agent-based models (ABMs) to examine public health outcomes and policy interventions associated with uneven urban food environments. While providing keen insights about the role of ABMs in studying urban food environments, these studies underutilize real-world data on individual behavior in their models. This study provides a unique contribution to the ABM and food access literature by utilizing survey data to develop an empirically-rich spatially-explicit ABM of food access. This model is used to simulate and scrutinize individual travel behavior associated with accessing food in low-income neighborhoods experiencing disinvestment in Detroit (Michigan), U.S. In particular, the relationship between trip frequencies, mode of travel, store choice, and distances traveled among individuals grouped into strata based on selected sociodemographic characteristics, including household income and age, is examined. Results reveal a diversified picture of not only how income and age shape food shopping travel but also the different thresholds of tolerance for non-motorized travel to stores. Younger and poorer population subgroups have a higher propensity to utilize non-motorized travel for shopping than older and wealthier subgroups. While all groups tend to travel considerable distances outside their immediate local food environment, different sociodemographic groups maintain unique spatial patterns of grocery-shopping behavior throughout the city and the suburbs. Overall, these results challenge foundational tenets in urban planning and design, regarding the specific characteristics necessary in the built environment to facilitate accessibility to urban amenities, such as grocery stores. In neighborhoods experiencing disinvestment, sociodemographic conditions play a more important role than the built environment in shaping food accessibility and ultimately travel behavior.

Public transit and depression among older adults: using agent-based models to examine plausible impacts of a free bus policy

BACKGROUND

Daily transport is associated with mental health. A free bus policy (FBP) may be effective in promoting the use of public transit in older adults and be associated with reductions in depressive symptoms.

METHODS

We developed an agent-based model and grounded it using empirical data from England to examine the impact of an FBP on public transit use and depression among older adults. We also used the model to explore whether the impact of the FBP bus use and depression is modified by the type of income segregation or by simultaneous efforts to improve attitudes towards the bus, to reduce waiting times or to increase the cost of driving via parking fees or fuel price.

RESULTS

Our model suggests that improving attitudes towards the bus (eg, campaigns that promote bus use) could enhance the effect of the FBP, especially for those in proximity to public transit. Reducing wait times could also significantly magnify FPB impacts, especially in those who live in proximity to public transit. Contrary to expectation, neither fuel costs nor parking fees significantly enhanced the impact of the FBP. The impact of improving attitudes towards the bus and increasing bus frequency was more pronounced in the lower-income groups in an income segregation scenario in which destination and public transit are denser in the city centre.

CONCLUSION

Our results suggest that the beneficial mental health effects of an FBP for older adults can be magnified when combined with initiatives that reduce bus waiting times and increased spatial access to transit.