How can agent-based modelling provide new insights into the impact of minimum unit pricing in Scotland?

In recent years we have gained insight into the impact of minimum unit pricing (MUP)-a legal floor price below which a given volume of alcohol cannot be sold-on population-level reductions in alcohol sales, consumption and harm. However, several questions remain unanswered including how individual-level purchasing changes impact the local economy (e.g., balance between on-licence and off-licence outlets), lead to long-term population-level trends (e.g., youth drinking) and social harms (e.g., violence). Agent-based modelling captures heterogeneity, emergence, feedback loops and adaptive and dynamic features, which provides an opportunity to understand the nuanced effects of MUP. Agent-based models (ABM) simulate heterogeneous agents (e.g., individuals, organisations) often situated in space and time that interact with other agents and/or with their environment, allowing us to identify the mechanisms underlying social phenomena. ABMs are particularly useful for theory development, and testing and simulating the impacts of policies and interventions. We illustrate how ABMs could be applied to generate novel insights and provide best estimates of social network effects, and changes in purchasing behaviour and social harms, due to the implementation of MUP. ABMs like other modelling approaches can simulate alternative implementations of MUP (e.g., policy intensity [£0.50, £0.60] or spatial scales [local, national]) but can also provide an understanding of the potential impact of MUP on different population groups (e.g., alcohol exposure of young people who are not yet drinking). Using ABMs to understand the impact of MUP would provide new insights to complement those from traditional epidemiological and other modelling methods.

Commentary on Antosz et al. (2023): The role of macro-micro-macro frameworks and critical realism in agent-based modelling.

Antosz and colleagues' review of the role of theory in agent-based modelling (ABM) makes important recommendations for modelling practitioners. However, macro-micro-macro frameworks are not necessarily as reliant on existing theory as the review suggests. Adopting a critical realist perspective to ABM design would help to deliver the recommendations, within which macro-micro-macro frameworks can play an important enabling role.

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.

Can Social Norms Explain Long-Term Trends in Alcohol Use? Insights from Inverse Generative Social Science.

Social psychological theory posits entities and mechanisms that attempt to explain observable differences in behavior. For example, dual process theory suggests that an agent's behavior is influenced by intentional (arising from reasoning involving attitudes and perceived norms) and unintentional (i.e., habitual) processes. In order to pass the generative sufficiency test as an explanation of alcohol use, we argue that the theory should be able to explain notable patterns in alcohol use that exist in the population, e.g., the distinct differences in drinking prevalence and average quantities consumed by males and females. In this study, we further develop and apply inverse generative social science (iGSS) methods to an existing agent-based model of dual process theory of alcohol use. Using iGSS, implemented within a multi-objective grammar-based genetic program, we search through the space of model structures to identify whether a single parsimonious model can best explain both male and female drinking, or whether separate and more complex models are needed. Focusing on alcohol use trends in New York State, we identify an interpretable model structure that achieves high goodness-of-fit for both male and female drinking patterns simultaneously, and which also validates successfully against reserved trend data. This structure offers a novel interpretation of the role of norms in formulating drinking intentions, but the structure's theoretical validity is questioned by its suggestion that individuals with low autonomy would act against perceived descriptive norms. Improved evidence on the distribution of autonomy in the population is needed to understand whether this finding is substantive or is a modeling artefact.

Guidance on the use of complex systems models for economic evaluations of public health interventions.

To help health economic modelers respond to demands for greater use of complex systems models in public health. To propose identifiable features of such models and support researchers to plan public health modeling projects using these models. A working group of experts in complex systems modeling and economic evaluation was brought together to develop and jointly write guidance for the use of complex systems models for health economic analysis. The content of workshops was informed by a scoping review. A public health complex systems model for economic evaluation is defined as a quantitative, dynamic, non-linear model that incorporates feedback and interactions among model elements, in order to capture emergent outcomes and estimate health, economic and potentially other consequences to inform public policies. The guidance covers: when complex systems modeling is needed; principles for designing a complex systems model; and how to choose an appropriate modeling technique. This paper provides a definition to identify and characterize complex systems models for economic evaluations and proposes guidance on key aspects of the process for health economics analysis. This document will support the development of complex systems models, with impact on public health systems policy and decision making.

Simulation of Alcohol Control Policies for Health Equity (SIMAH) Project: Study Design and First Results.

Since about 2010, life expectancy at birth in the United States has stagnated and begun to decline, with concurrent increases in the socioeconomic divide in life expectancy. The Simulation of Alcohol Control Policies for Health Equity (SIMAH) Project uses a novel microsimulation approach to investigate the extent to which alcohol use, socioeconomic status (SES), and race/ethnicity contribute to unequal developments in US life expectancy and how alcohol control interventions could reduce such inequalities. Representative, secondary data from several sources will be integrated into one coherent, dynamic microsimulation to model life-course changes in SES and alcohol use and cause-specific mortality attributable to alcohol use by SES, race/ethnicity, age, and sex. Markov models will be used to inform transition intensities between levels of SES and drinking patterns. The model will be used to compare a baseline scenario with multiple counterfactual intervention scenarios. The preliminary results indicate that the crucial microsimulation component provides a good fit to observed demographic changes in the population, providing a robust baseline model for further simulation work. By demonstrating the feasibility of this novel approach, the SIMAH Project promises to offer superior integration of relevant empirical evidence to inform public health policy for a more equitable future.

Behavioral stability of alcohol consumption and socio-demographic correlates of change among a nationally representative cohort of US adults.

AIMS: To estimate the probability of transitioning between different categories of alcohol use (drinking states) among a nationally representative cohort of United States (US) adults and to identify the effects of socio-demographic characteristics on those transitions. DESIGN, SETTING AND PARTICIPANTS: Secondary analysis of data from the National Epidemiologic Survey of Alcohol and Related Conditions (NESARC), a prospective cohort study conducted in 2001-02 and 2004-05; a US nation-wide, population-based study. Participants included 34 165 adults (mean age = 45.1 years, standard deviation = 17.3; 52% women). MEASUREMENTS: Alcohol use was self-reported and categorized based on the grams consumed per day: (1) non-drinker (no drinks in past 12 months), (2) category I (women = ≤ 20; men = ≤ 40), (3) category II (women = 21-40; men = 41-60) and (4) category III (women = ≥ 41; men = ≥ 61). Multi-state Markov models estimated the probability of transitioning between drinking states, conditioned on age, sex, race/ethnicity and educational attainment. Analyses were repeated with alcohol use categorized based on the frequency of heavy episodic drinking. FINDINGS: The highest transition probabilities were observed for staying in the same state; after 1 year, the probability of remaining in the same state was 90.1% [95% confidence interval (CI) = 89.7%, 90.5%] for non-drinkers, 90.2% (95% CI = 89.9%, 90.5%) for category I, 31.8% (95% CI = 29.7, 33.9%) category II and 52.2% (95% CI = 46.0, 58.5%) for category III. Women, older adults, and non-Hispanic Other adults were less likely to transition between drinking states, including transitions to lower use. Adults with lower educational attainment were more likely to transition between drinking states; however, they were also less likely to transition out of the 'weekly HED' category. Black adults were more likely to transition into or stay in higher use categories, whereas Hispanic/Latinx adults were largely similar to White adults. CONCLUSIONS: In this study of alcohol transition probabilities, some demographic subgroups appeared more likely to transition into or persist in higher alcohol consumption states.

A synthetic population dataset for estimating small area health and socio-economic outcomes in Great Britain.

In order to understand the health outcomes for distinct sub-groups of the population or across different geographies, it is advantageous to be able to build bespoke groupings from individual level data. Individuals possess distinct characteristics, exhibit distinct behaviours and accumulate their own unique history of exposure or experiences. However, in most disciplines, not least public health, there is a lack of individual level data available outside of secure settings, especially covering large portions of the population. This paper provides detail on the creation of a synthetic micro dataset for individuals in Great Britain who have detailed attributes which can be used to model a wide range of health and other outcomes. These attributes are constructed from a range of sources including the United Kingdom Census, survey and administrative datasets. It provides a rationale for the need for this synthetic population, discusses methods for creating this dataset and provides some example results of different attribute distributions for distinct sub-population groups and over different geographical areas.

Improved estimates for individual and population-level alcohol use in the United States, 1984-2020.

Aims: While nationally representative alcohol surveys are a mainstay of public health monitoring, they underestimate consumption at the population level. This paper demonstrates how to adjust individual-level survey data using aggregated alcohol per capita (APC) data for improved individual- and population-level consumption estimates. Design and Methods: For the period 1984-2020, data on self-reported alcohol consumption in the past 30 days were taken from the Behavioral Risk Factor Surveillance System (BRFSS) involving participants (18+ years) in the United States (US). Monthly abstainers were reallocated into lifetime abstainers, former drinkers, and 12-month drinkers using the 2005 National Alcohol Survey data. To correct for under-coverage of alcohol use, we triangulated APC and survey data by upshifting quantity (average grams/day) and frequency (drinking days/week) of alcohol use based on national- and state-level APC data. Results were provided for the US as a whole and for selected states to represent different drinking patterns. Findings: The corrections described above resulted in improved correspondence between survey and APC data. Following our procedure, national estimates of alcohol quantity increased from 45% to 77% of APC estimates. Both quantity and frequency of alcohol use were upshifted; by upshifting to 90% of APC, we were able to fit trends and distributions in APC patterns for individual states and the US. Conclusions: An individual-level dataset which more accurately reflects the alcohol use of US citizens was achieved. This dataset will be invaluable as a research tool and for the planning and evaluation of alcohol control policies for the US. The methodology described can also be used to adjust individual-level alcohol survey data in other geographical settings.

Causal mechanisms proposed for the alcohol harm paradox-a systematic review.

BACKGROUND AND AIMS: The alcohol harm paradox (AHP) posits that disadvantaged groups suffer from higher rates of alcohol-related harm compared with advantaged groups, despite reporting similar or lower levels of consumption on average. The causes of this relationship remain unclear. This study aimed to identify explanations proposed for the AHP. Secondary aims were to review the existing evidence for those explanations and investigate whether authors linked explanations to one another. METHODS: This was a systematic review. We searched MEDLINE (1946-January 2021), EMBASE (1974-January 2021) and PsycINFO (1967-January 2021), supplemented with manual searching of grey literature. Included papers either explored the causes of the AHP or investigated the relationship between alcohol consumption, alcohol-related harm and socio-economic position. Papers were set in Organization for Economic Cooperation and Development high-income countries. Explanations extracted for analysis could be evidenced in the empirical results or suggested by researchers in their narrative. Inductive thematic analysis was applied to group explanations. RESULTS: Seventy-nine papers met the inclusion criteria and initial coding revealed that these papers contained 41 distinct explanations for the AHP. Following inductive thematic analysis, these explanations were grouped into 16 themes within six broad domains: individual, life-style, contextual, disadvantage, upstream and artefactual. Explanations related to risk behaviours, which fitted within the life-style domain, were the most frequently proposed (n = 51) and analysed (n = 21). CONCLUSIONS: While there are many potential explanations for the alcohol harm paradox, most research focuses on risk behaviours while other explanations lack empirical testing.

An integrated dual process simulation model of alcohol use behaviours in individuals, with application to US population-level consumption, 1984-2012.

INTRODUCTION: The Theory of Planned Behaviour (TPB) describes how attitudes, norms and perceived behavioural control guide health behaviour, including alcohol consumption. Dual Process Theories (DPT) suggest that alongside these reasoned pathways, behaviour is influenced by automatic processes that are determined by the frequency of engagement in the health behaviour in the past. We present a computational model integrating TPB and DPT to determine drinking decisions for simulated individuals. We explore whether this model can reproduce historical patterns in US population alcohol use and simulate a hypothetical scenario, "Dry January", to demonstrate the utility of the model for appraising the impact of policy interventions on population alcohol use. METHOD: Constructs from the TPB pathway were computed using equations from an existing individual-level dynamic simulation model of alcohol use. The DPT pathway was initialised by simulating individuals' past drinking using data from a large US survey. Individuals in the model were from a US population microsimulation that accounts for births, deaths and migration (1984-2015). On each modelled day, for each individual, we calculated standard drinks consumed using the TPB or DPT pathway. In each year we computed total population alcohol use prevalence, frequency and quantity. The model was calibrated to alcohol use data from the Behavioral Risk Factor Surveillance System (1984-2004). RESULTS: The model was a good fit to prevalence and frequency but a poorer fit to quantity of alcohol consumption, particularly in males. Simulating Dry January in each year led to a small to moderate reduction in annual population drinking. CONCLUSION: This study provides further evidence, at the whole population level, that a combination of reasoned and implicit processes are important for alcohol use. Alcohol misuse interventions should target both processes. The integrated TPB-DPT simulation model is a useful tool for estimating changes in alcohol consumption following hypothetical population interventions.

Causality and initiation of alcohol control policy. A response to Allamani.

In a recent commentary, Allamani asked how one can establish causality in epidemiological research, and specifically about causality as it relates to alcohol control policy. Epidemiology customarily uses a sufficient-component cause model, where a sufficient cause for an outcome is determined by a set of minimal conditions and events that inevitably produce the stated outcome. While this model is theoretically clear, its operationalisation often involves probabilistic elements. Recent advances in agent-based modelling may improve operationalisation. The implications for alcohol control policy from this model are straightforward: the so-called alcohol-attributable fraction denotes the cases of morbidity or mortality which would not have happened in the absence of alcohol use.

Beyond Behaviour: How Health Inequality Theory Can Enhance Our Understanding of the 'Alcohol-Harm Paradox'.

There are large socioeconomic inequalities in alcohol-related harm. The alcohol harm paradox (AHP) is the consistent finding that lower socioeconomic groups consume the same or less as higher socioeconomic groups yet experience greater rates of harm. To date, alcohol researchers have predominantly taken an individualised behavioural approach to understand the AHP. This paper calls for a new approach which draws on theories of health inequality, specifically the social determinants of health, fundamental cause theory, political economy of health and eco-social models. These theories consist of several interwoven causal mechanisms, including genetic inheritance, the role of social networks, the unequal availability of wealth and other resources, the psychosocial experience of lower socioeconomic position, and the accumulation of these experiences over time. To date, research exploring the causes of the AHP has often lacked clear theoretical underpinning. Drawing on these theoretical approaches in alcohol research would not only address this gap but would also result in a structured effort to identify the causes of the AHP. Given the present lack of clear evidence in favour of any specific theory, it is difficult to conclude whether one theory should take primacy in future research efforts. However, drawing on any of these theories would shift how we think about the causes of the paradox, from health behaviour in isolation to the wider context of complex interacting mechanisms between individuals and their environment. Meanwhile, computer simulations have the potential to test the competing theoretical perspectives, both in the abstract and empirically via synthesis of the disparate existing evidence base. Overall, making greater use of existing theoretical frameworks in alcohol epidemiology would offer novel insights into the AHP and generate knowledge of how to intervene to mitigate inequalities in alcohol-related harm.

Using Multi-objective Grammar-based Genetic Programming to Integrate Multiple Social Theories in Agent-based Modeling.

Different theoretical mechanisms have been proposed for explaining complex social phenomena. For example, explanations for observed trends in population alcohol use have been postulated based on norm theory, role theory, and others. Many mechanism-based models of phenomena attempt to translate a single theory into a simulation model. However, single theories often only represent a partial explanation for the phenomenon. The potential of integrating theories together, computationally, represents a promising way of improving the explanatory capability of generative social science. This paper presents a framework for such integrative model discovery, based on multi-objective grammar-based genetic programming (MOGGP). The framework is demonstrated using two separate theory-driven models of alcohol use dynamics based on norm theory and role theory. The proposed integration considers how the sequence of decisions to consume the next drink in a drinking occasion may be influenced by factors from the different theories. A new grammar is constructed based on this integration. Results of the MOGGP model discovery process find new hybrid models that outperform the existing single-theory models and the baseline hybrid model. Future work should consider and further refine the role of domain experts in defining the meaningfulness of models identified by MOGGP.

The role of alcohol use in the aetiology and progression of liver disease: A narrative review and a quantification.

ISSUES: Alcohol use has been shown to impact on various forms of liver disease, not restricted to alcoholic liver disease. APPROACH: We developed a conceptual framework based on a narrative review of the literature to identify causal associations between alcohol use and various forms of liver disease including the complex interactions of alcohol with other major risk factors. Based on this framework, we estimate the identified relations for 2017 for the USA. KEY FINDINGS: The following pathways were identified and modelled for the USA for the year 2017. Alcohol use caused 35 200 (95% uncertainty interval 32 800-37 800) incident cases of alcoholic liver cirrhosis. There were 1700 (uncertainty interval 1100-2500) acute hepatitis B and C virus (HBV and HCV) infections attributable to heavy-drinking occasions, and 14 000 (uncertainty interval 5900-19 500) chronic HBV and 1700 (uncertainty interval 700-2400) chronic HCV infections due to heavy alcohol use interfering with spontaneous clearance. Alcohol use and its interactions with other risk factors (HBV, HCV, obesity) led to 54 500 (uncertainty interval 50 900-58 400) new cases of liver cirrhosis. In addition, alcohol use caused 6600 (uncertainty interval 4200-9300) liver cancer deaths and 40 700 (uncertainty interval 36 600-44 600) liver cirrhosis deaths. IMPLICATIONS: Alcohol use causes a substantial number of incident cases and deaths from chronic liver disease, often in interaction with other risk factors. CONCLUSION: This additional disease burden is not reflected in the current alcoholic liver disease categories. Clinical work and prevention policies need to take this into consideration.

Impact of body mass and alcohol consumption on all-cause and liver mortality in 240 000 adults in the United States.

INTRODUCTION: Nationally representative studies of the combined impact of drinking and body mass (BMI) on mortality outcomes are unavailable. We investigate whether both act together to elevate risk of all-cause or liver mortality. METHODS: We obtained self-reported histories of drinking and BMI from 129 098 women (mean age 47.2 years) and 102 568 men (mean age 45.6 years) ≥18 years interviewed from 1997 to 2004 in the National Health Interview Survey and related these data to the deaths that occurred by 31 December 2006 (women = 8486; men = 7819 deaths). Death hazards among current drinkers in different BMI groups were adjusted for age, education, race and smoking. RESULTS: Obese (≥30 kg m-2 ) adults with consumption of >40 g day-1 (women) or >60 g day-1 (men) pure ethanol were at risk of increased mortality from all-cause and chronic liver disease (P trend <0.0001). For heavy drinkers with BMI ≥30 kg m-2 , each 5 kg m-2 higher BMI was associated with an elevated all-cause mortality in men (hazard ratios 1.27, 95% confidence interval [CI]: 1.16-1.40) and women (1.12, [1.02-1.24]). The excess risk due to interaction was more pronounced in men (7.30, [3.60-11.00]) than women (2.90, [0.50-5.30]). DISCUSSION AND CONCLUSIONS: Obesity and excess alcohol are both related to all-cause and liver mortality-the latter with evidence of a supra-additive interaction between the risk factors. The presence of both factors in the same population and their impact should inform treatment, public health policies and research.

Multiobjective Genetic Programming Can Improve the Explanatory Capabilities of Mechanism-Based Models of Social Systems.

The generative approach to social science, in which agent-based simulations (or other complex systems models) are executed to reproduce a known social phenomenon, is an important tool for realist explanation. However, a generative model, when suitably calibrated and validated using empirical data, represents just one viable candidate set of entities and mechanisms. The model only partially addresses the needs of an abductive reasoning process - specifically it does not provide insight into other viable sets of entities or mechanisms, nor suggest which of these are fundamentally constitutive for the phenomenon to exist. In this paper, we propose a new model discovery framework that more fully captures the needs of realist explanation. The framework exploits the implicit ontology of an existing human-built generative model to propose and test a plurality of new candidate model structures. Genetic programming is used to automate this search process. A multi-objective approach is used, which enables multiple perspectives on the value of any particular generative model - such as goodness-of-fit, parsimony, and interpretability - to be represented simultaneously. We demonstrate this new framework using a complex systems modeling case study of change and stasis in societal alcohol use patterns in the US over the period 1980-2010. The framework is successful in identifying three competing explanations of these alcohol use patterns, using novel integrations of social role theory not previously considered by the human modeler. Practitioners in complex systems modeling should use model discovery to improve the explanatory utility of the generative approach to realist social science.

A software architecture for mechanism-based social systems modelling in agent-based simulation models.

This paper introduces the MBSSM (Mechanism-Based Social Systems Modelling) software architecture that is designed for expressing mechanisms of social theories with individual behaviour components in a unified way and implementing these mechanisms in an agent-based simulation model. The MBSSM architecture is based on a middle-range theory approach most recently expounded by analytical sociology and is designed in the object-oriented programming paradigm with Unified Modelling Language diagrams. This paper presents two worked examples of using the architecture for modelling individual behaviour mechanisms that give rise to the dynamics of population-level alcohol use: a single-theory model of norm theory and a multi-theory model that combines norm theory with role theory. The MBSSM architecture provides a computational environment within which theories based on social mechanisms can be represented, compared, and integrated. The architecture plays a fundamental enabling role within a wider simulation model-based framework of abductive reasoning in which families of theories are tested for their ability to explain concrete social phenomena.

Expanding attributable fraction applications to outcomes wholly attributable to a risk factor.

The problem central to this document is the estimation of change in disease attributable to an epidemiological exposure variable that stems from a change in the distribution of that variable. We require that both disease and exposure are quantifiable as real numbers, and then ask how to estimate the fraction of disease attributable to exposure, producing the general attributable fraction methodology. After the mathematical framework is in place, we explore the implications of a disease that is wholly attributable to a given risk factor, demonstrate why standard applications of the attributable fractions do not extend, and present general methodological considerations for this case. Finally, we demonstrate the methodology using the example of alcoholic psychoses.