Counterfactual Approach Research Articles

Effectiveness of a community health worker-led case management programme to improve outcomes for people with psychotic disorders in Thailand: a one-year prospective cohort study

BackgroundIntensive case management (ICM) programmes for psychotic patients are effective in improving outcomes, but often unfeasible in resource-poor settings, as they typically require extensive human resources and expertise. We developed and evaluated the effectiveness of a less intensive case management program (LICM), led by community health workers, on one-year social functioning and service use.MethodsA prospective cohort study was conducted on patients aged 18 and above residing in a hospital catchment area. Outcomes were compared between LICM (n = 64) and non-LICM participants (n = 485). A counterfactual framework approach was applied to assess causal effects of the LICM on outcomes. The programme effectiveness was analyzed by augmented-inverse probability of treatment weighting (AIPW) to estimate potential outcome mean (POM) and average treatment effect (ATE). Outcomes were employment status and use of emergency, inpatient and outpatient services. Analyses were stratified by the number of previous psychotic relapse (≤ 1, > 1) to assess heterogeneity of treatment effect on those in an early and later stages of psychotic illness.ResultsIn the early-stage cohort, the likelihood of being employed at one year post-baseline was significantly greater in LICM participants than non-LICM participants (ATE 0.10, 95%CI 0.05–0.14, p < 0.001), whereas service use of all types, except outpatient, was not significantly different between the two groups. In the later-stage cohort, the likelihoods of employment between the two groups at post-baseline were similar (ATE -0.02, 95%CI -0.19–0.15, p = 0.826), whereas service use of all types was significantly higher in LICM participants.ConclusionLICM in a setting where community mental services are scarce may benefit those at an early stage of psychotic illness, by leading to better social functioning and no higher use of unscheduled services at the end of the programme, possibly through their better prognosis and medication adherence. A more intensive case management model may be appropriate for those in a later stage of the illness.

Self-efficacy, burnout and academic success in nursing students: A counterfactual mediation analysis.

To test the mediating role of burnout in the relationship between self-efficacy and academic success in nursing students. This was a cross-sectional secondary analysis of longitudinal research aimed at exploring the academic success of nursing students. We enrolled a convenience sample of nursing students attending 21 Italian baccalaureate nursing degree programmes. Data were collected from 2014 to the 2018-2019 academic year. We analysed the wave of data collected during the third year of the programme. The measurements were the Academic Nurse Self-Efficacy scale (ANSEs) and the Scale of Work Burnout (SWEBO). A mediation analysis with a counterfactual approach was performed. The participants (n=556) had a median age of 20years (IQR 19-22) and the majority was female [70.5% (392/556)]. Academic success was reached by 51.97% (289/556) of the students. The findings reveal a significant direct effect of self-efficacy on academic success and also a significant indirect effect of this relationship via burnout. The mechanism by which self-efficacy influences academic performance in nursing students is more complex than a simple direct relationship. Universities should consider screening students for variables affecting academic success to decrease academic costs and increase ranking systems classification. Nurses' professors should be engaged in strategies to promote self-efficacy. The area of intervention should be structured both during classes and internships, in tandem with the mentor and the clinical nurse.

It's giving me the blues: A fixed-effects and g-formula approach to understanding job insecurity, sleep disturbances, and major depression

Research suggests that work-related factors like job insecurity increases the risk of major depression (MD), although it is unclear whether the association is causal. Research further suggests that job insecurity increases sleep disturbances, which is also a risk factor for MD. Based on current knowledge, it is possible that job insecurity operates through sleep disturbances to affect MD, but this pathway has not been examined in the literature. The current study extends the literature by using two complementary, counterfactual approaches (i.e., random- and fixed-effects regression and a mediational g-formula) to examine whether job insecurity causes MD and whether sleep disturbances mediate the relationship. A methodological triangulation approach allowed us to adjust for unobserved and intermediate confounding, which has not been addressed in prior research. Findings suggest that the relationship between job insecurity and MD is primarily direct, that hypothetically intervening on job insecurity (in our g-formula) would reduce MD by approximately 10% at the population level, and this relationship operates via sleep disturbances to some degree. However, the indirect pathway had a high degree of uncertainty.

Long-term labour market and economic consequences of school exclusions in England: Evidence from two counterfactual approaches.

BackgroundPrevious research suggests that school exclusion during childhood is a precursor to social exclusion in adulthood. Past literature on the consequences of school exclusion is, however, scarce and mainly focused on short‐term outcomes such as educational attainment, delinquency, and mental health in early adolescence. Moreover, this evidence is based primarily on descriptive and correlational analysis, whereas robust causal evidence is required to best inform policy.AimsWe aimed to estimate the mid‐to‐long‐term impact of school exclusion on labour market and economic outcomes.SampleThe sample included 6,632 young people who at the age of 25/26 in the year 2015 participated in the Next Steps survey of whom 86 were expelled from school and 711 were suspended between the ages of 13/14 and 16/17.MethodUsing high quality existing longitudinal data, we utilized four approaches to evaluate the impact of school exclusion: logistic regression‐adjustment models, propensity score matching, school fixed‐effects analysis, and inverse propensity weighting. The latter two counterfactual approaches were used to estimate causal effects.ResultsWe found that school exclusion increased the risk of becoming NEET at the age of 19/20, and then remaining economically inactive at the age of 25/26, as well as experiencing higher unemployment risk and earning lower wages also at the age of 25/26.ConclusionSchool exclusion has pervasive negative effects into adulthood. Policy interventions should focus on both prevention and mitigating its negative effects. Interventions aimed at re‐integrating excluded individuals into education or vocational training could be key in reducing the risk of poor socio‐economic outcomes and social exclusion.

Getting counterfactuals right: the perspective of the causal reasoner

This paper aims to bridge philosophical and psychological research on causation, counterfactual thought, and the problem of backtracking. Counterfactual approaches to causation such as that by Lewis have ruled out backtracking, while on prominent models of causal inference interventionist counterfactuals do not backtrack. However, on various formal models, certain backtracking counterfactuals end up being true, and psychological evidence shows that people do sometimes backtrack when answering counterfactual questions in causal contexts. On the basis of psychological research, I argue that while ordinarily both kinds of counterfactuals may be employed, non-backtracking counterfactuals are more easily used in causal inference because they are consistent with temporal order information embedded in the mental simulation heuristic, and they match reasoners’ experience of causation. While this approach is incompatible with the ambitions of counterfactual theories that seek to establish the non-backtracking interpretation as the only legitimate one, it can provide support for perspectival views on causation and open further inquiry on the functions of causal and counterfactual thought in the context of causal models.

Counterfactual inference with latent variable and its application in mental health care.

This paper deals with the problem of modeling counterfactual reasoning in scenarios where, apart from the observed endogenous variables, we have a latent variable that affects the outcomes and, consequently, the results of counterfactuals queries. This is a common setup in healthcare problems, including mental health. We propose a new framework where the aforementioned problem is modeled as a multivariate regression and the counterfactual model accounts for both observed and a latent variable, where the latter represents what we call the patient individuality factor (upvarphi ). In mental health, focusing on individuals is paramount, as past experiences can change how people see or deal with situations, but individuality cannot be directly measured. To the best of our knowledge, this is the first counterfactual approach that considers both observational and latent variables to provide deterministic answers to counterfactual queries, such as: what if I change the social support of a patient, to what extent can I change his/her anxiety? The framework combines concepts from deep representation learning and causal inference to infer the value of upvarphi and capture both non-linear and multiplicative effects of causal variables. Experiments are performed with both synthetic and real-world datasets, where we predict how changes in people’s actions may lead to different outcomes in terms of symptoms of mental illness and quality of life. Results show the model learns the individually factor with errors lower than 0.05 and answers counterfactual queries that are supported by the medical literature. The model has the potential to recommend small changes in people’s lives that may completely change their relationship with mental illness.

Does vitamin A rich orange-fleshed sweetpotato adoption improve household level diet diversity? Evidence from Ghana and Nigeria

Promoting the consumption of locally available food crops such as orange-fleshed sweetpotato (OFSP) that are rich in beta carotene, a precursor for vitamin A, has been shown to be a highly effective means of fighting against vitamin A deficiency at the community level when combined with nutrition education. After a three-year intervention promoting OFSP uptake using different market-led approaches, a structured survey was conducted among 204 participant and 422 non-participant households in Ghana and Nigeria in 2017. In this study, we test whether the adoption of OFSP has a significant effect on dietary diversity, using a counterfactual approach based on the conditional independence-based estimators of average treatment effect techniques. Results found that household, young child and woman’s dietary diversity scores increased when households adopted OFSP varieties. However, the positive effects of OFSP adoption on household dietary diversity only occurred in the Ghanaian sub-sample, not the Nigerian. Policymakers and development support partners should ensure institutional support in terms of intensive campaigns to sensitize farmers and their families about the benefits of these crops in order to create and sustain demand for biofortified foods as a means of fighting micronutrient malnutrition.

Structural Racism and Quantitative Causal Inference: A Life Course Mediation Framework for Decomposing Racial Health Disparities.

Quantitative studies of racial health disparities often use static measures of self-reported race and conventional regression estimators, which critics argue is inconsistent with social-constructivist theories of race, racialization, and racism. We demonstrate an alternative counterfactual approach to explain how multiple racialized systems dynamically shape health over time, examining racial inequities in cardiometabolic risk in the National Longitudinal Study of Adolescent to Adult Health. This framework accounts for the dynamics of time-varying confounding and mediation that is required in operationalizing a "race" variable as part of a social process (racism) rather than a separable, individual characteristic. We decompose the observed disparity into three types of effects: a controlled direct effect ("unobserved racism"), proportions attributable to interaction ("racial discrimination"), and pure indirect effects ("emergent discrimination"). We discuss the limitations of counterfactual approaches while highlighting how they can be combined with critical theories to quantify how interlocking systems produce racial health inequities.

A potential outcomes approach to defining and estimating gestational age-specific exposure effects during pregnancy.

Many studies seek to evaluate the effects of potentially harmful pregnancy exposures during specific gestational periods. We consider an observational pregnancy cohort where pregnant individuals can initiate medication usage or become exposed to a drug at various times during their pregnancy. An important statistical challenge involves how to define and estimate exposure effects when pregnancy loss or delivery can occur over time. Without proper consideration, the results of standard analysis may be vulnerable to selection bias, immortal time-bias, and time-dependent confounding. In this study, we apply the “target trials” framework of Hernán and Robins in order to define effects based on the counterfactual approach often used in causal inference. This effect is defined relative to a hypothetical randomized trial of timed pregnancy exposures where delivery may precede and thus potentially interrupt exposure initiation. We describe specific implementations of inverse probability weighting, G-computation, and Targeted Maximum Likelihood Estimation to estimate the effects of interest. We demonstrate the performance of all estimators using simulated data and show that a standard implementation of inverse probability weighting is biased. We then apply our proposed methods to a pharmacoepidemiology study to evaluate the potentially time-dependent effect of exposure to inhaled corticosteroids on birthweight in pregnant people with mild asthma.

When less conditioning provides better estimates: overcontrol and endogenous selection biases in research on intergenerational mobility

The counterfactual approach to causality has become the dominant approach to understand causality in contemporary social science research. Whilst most sociologists are aware that unobserved, confounding variables may bias the estimates of causal effects (omitted variable bias), the threats of overcontrol and endogenous selection biases are less well known. In particular, widely used practices in research on intergenerational mobility are affected by these biases. I review four of these practices from the viewpoint of the counterfactual approach to causality and show why overcontrol and endogenous selection biases arise when these practices are implemented. I use data from the German Socio-Economic Panel Study (SOEP) to demonstrate the practical consequences of these biases for conclusions about intergenerational mobility. I conclude that future research on intergenerational mobility should reflect more upon the possibilities of bias introduced by conditioning on variables.

Cesarean section and body mass index in children: is there a causal effect?

Obesity is considered a global public health problem. Cesarean section has been associated with high body mass index (BMI) and increased obesity throughout life. However, this association has been challenged by some studies. This study aims to assess the causal effect of cesarean section on the BMI of children aged 1-3 years. This is a cohort study of 2,181 children aged 1-3 years, born in 2010, obtained from the BRISA Birth Cohort, in São Luís, state of Maranhão, Brazil. Sociodemographic variables, maternal characteristics, type of childbirth, morbidity, anthropometric measurements, and BMI were assessed. Marginal structural models with a counterfactual approach were used to check the causal effect of the type of childbirth on obesity, weighted by the inverse probability of selection and exposure. Out of the 2,181 children assessed (52% female), 50.6% were born by cesarean section, 5.9% of the newborn infants were large for gestational age, and 10.7% of them had excess weight. No causal effect of cesarean section on BMI was observed (coefficient = -0.004; 95%CI: -0.136; 0.127; p = 0.948). Cesarean section did not have a causal effect on the BMI of children aged 1-3 years.

FCE: Feedback Based Counterfactual Explanations for Explainable AI

Artificial Intelligence provides accurate predictions for critical applications (e.g., healthcare, finance), but lacks the ability to explain its internal mechanism in most applications which require high interaction with humans. Even if many studies analyze machine learning models and their learning behaviour and eventually provide an interpretation of the inner mechanics of these models, these studies often entail a simpler surrogate model generate explanations by producing a piece of interpretable information such as feature scores. The crucial caveat against these studies is the lack of human involvement in the design and evaluation of explanations, consequently giving rise to trust issues and lack of acceptance and understanding. To this end, we address this limitation by involving humans in the counterfactual explanation generation process which is enriched with user feedback and thus enhancing the automated explanations which are better aligned with user expectations. In this paper, we propose a user feedback based counterfactual explanation approach for explainable Artificial Intelligence. In our work, we utilize feedback in two ways: first, feedback to customize the explanations locally that helps in providing the neighbourhood to discern the feasible explanations; and second, to evaluate the generated explanations.

Adjusting Expected Deaths for Mortality Displacement During the COVID-19 Pandemic: A Model Based Counterfactual Approach at the Level of Individuals

Adjusting Expected Deaths for Mortality Displacement During the COVID-19 Pandemic: A Model Based Counterfactual Approach at the Level of Individuals

A model-based study to estimate the health and economic impact of health technology assessment in Thailand.

Health technology assessment (HTA) plays a central role in the coverage and reimbursement decision-making process for public health expenditure in many countries, including Thailand. However, there have been few attempts to quantitatively understand the benefits of using HTA to inform resource allocation decisions. The objective of this research was to simulate the expected net monetary benefit (NMB) from using HTA-based decision criteria compared to a first-come, first-served (FCFS) approach using data from Thailand. A previously published simulation model was adapted to the Thai context which aimed to simulate the impact of using different decision-making criteria to adopt or reject health technologies for public reimbursement. Specifically, the simulation model provides a quantitative comparison between an HTA-based funding rule and a counterfactual (FCFS) funding rule to make decisions on which health technologies should be funded. The primary output of the model was the NMB of using HTA-based decision criteria compared to the counterfactual approach. The HTA-based decision rule in the model involved measuring incremental cost-effectiveness ratios against a cost-effectiveness threshold. The counterfactual decision rule was a FCFS (random) selection of health technologies. The HTA-based decision rule was associated with a greater NMB compared to the counterfactual. In the investigated analyses, the NMB ranged from THB24,238 million (USD725 million) to THB759,328 million (USD22,719 million). HTA-based decisions led to fewer costs, superior health outcomes (more quality-adjusted life-years). The results support the hypothesis that HTA can provide health and economic benefits by improving the efficiency of resource allocation decision making.

Evaluating the impact of COVID-19 on ex-vessel prices using time-series analysis

The spread of coronavirus disease 2019 (COVID-19) and subsequent lockdown measures have impacted economies and industries worldwide. The fisheries industry witnessed a sharp decline in demand and a slump in fish prices due to its dependence on the food service industry. It is important to quantitatively assess those fish species affected most and the extent of the pandemic’s impact on them, to take specific countermeasures. We propose a time-series analysis as an alternative to the current practice of using ad hoc year-on-year comparisons. Although the pandemic makes it difficult to construct a counterfactual approach due to the lack of an appropriate control group, we use time-series forecasting to simulate normal conditions using pre-pandemic data. In Tokyo, the unit price of fish species that were negatively impacted by the food services industry dropped by 12.65% to 14.64%, and by 26.08% to 28.22% after the declaration of a state of emergency. Seasonality, short weekly cycles, and short-term market trends are factors that affect the price of fish. Species-specific impact estimates related to the COVID-19 pandemic can allow policymakers to implement recovery measures in a more targeted and effective manner. The results of our analysis can increase fishers’ and policymakers’ awareness of the usefulness of economic analyses and incentivize them to release data to establish a system to accumulate and analyze data strategically for urgent and appropriate interventions in the fisheries industry.

Historical Injustice in Immigration Policy

The history of immigration policy is marked by the wrongful and discriminatory exclusion of certain groups of people. In this article, I argue that descendants of those who were wrongfully excluded have a pro tanto right to immigrate to the state in question as reparation. I begin by identifying the two main approaches theorists generally take to establish a claim for reparation: the inheritance approach and the counterfactual approach. In the first section, I argue that the inheritance approach does not offer a promising argument for reparations for descendants of those who were wrongfully excluded. In the second section, I argue that the counterfactual approach, by contrast, does. In the third section, I respond to the objection that this prima facie claim for reparation can be undermined by current circumstances. In the fourth section, I show why this reparation should be offered in the form of immigration rights.

COUNTERFACTUAL TEMPORAL MODEL OF CAUSAL RELATIONSHIPS FOR CONSTRUCTING EXPLANATIONS IN INTELLIGENT SYSTEMS

The subject of the research is the processes of constructing explanations based on causal relationships between states or actions of an intellectualsystem. An explanation is knowledge about the sequence of causes and effects that determine the process and result of an intelligent informationsystem. The aim of the work is to develop a counterfactual temporal model of cause-and-effect relationships as part of an explanation of the process offunctioning of an intelligent system in order to ensure the identification of causal dependencies based on the analysis of the logs of the behavior ofsuch a system. To achieve the stated goals, the following tasks are solved: determination of the temporal properties of the counterfactual description ofcause-and-effect relationships between actions or states of an intelligent information system; development of a temporal model of causal connections,taking into account both the facts of occurrence of events in the intellectual system, and the possibility of occurrence of events that do not affect theformation of the current decision. Conclusions. The structuring of the temporal properties of causal links for pairs of events that occur sequentially intime or have intermediate events is performed. Such relationships are represented by alternative causal relationships using the temporal operators"Next" and "Future", which allows realizing a counterfactual approach to the representation of causality. A counterfactual temporal model of causalrelationships is proposed, which determines deterministic causal relationships for pairs of consecutive events and pairs of events between which thereare other events, which determines the transitivity property of such dependencies and, accordingly, creates conditions for describing the sequence ofcauses and effects as part of the explanation in intelligent system with a given degree of detail The model provides the ability to determine cause-andeffect relationships, between which there are intermediate events that do not affect the final result of the intelligent information system.

The rise in youth employment? Impact evaluation of COVID-19 consequences

The COVID-19 pandemic has significant consequences in many areas and has largely contributed to rising unemployment in almost all countries. The situation is similar in Slovakia, where various degrees of shut-down measures have caused an inflow of newly unemployed people in all age groups. This article focuses on unemployed youth as a vulnerable and disadvantaged group of the working population, as starting a career during a pandemic is complicated or even impossible in some industries under these conditions. In this study, the real situation in Slovakia in 2020-2021 is compared with the development of the job market before the pandemic. The main contribution of the study is the quantification of the extent of the pandemic impact on unemployed youth, but also on other age groups; this evaluation can be considered very accurate given the counterfactual approach used. The results of this study can be used in practice to identify the most affected groups of the population and to implement policy measures aimed at mitigating the effects of the pandemic, and then to adjust the intensity and amount of allocated funds that will be needed to be spent to support the placement of young people into the labor market.

Are sex differences in the outcome of peritoneal dialysis explained by nurse assistance? A cohort study with data from the Registre de Dialyse Péritonéale de Langue Française.

We aimed to evaluate sex differences in peritoneal dialysis (PD) outcomes and to explore direct and indirect effects of nurse-assisted PD on outcomes. This was a retrospective study using data from the Registre de Dialyse Péritonéale de Langue Française of incident PD patients between 2005 and 2016. Cox proportional hazards modelling was used to analyse transfer to haemodialysis (HD), death, PD failure, peritonitis and renal transplantation. Mediation analyses with a counterfactual approach were carried out to evaluate natural direct and indirect effects of sex on transfer to HD and peritonitis, with nurse-assisted PD as a mediator a priori. Of the 14 659 patients included, there were 5970 females (41%) and 8689 males (59%). Women were more frequently treated by nurse-assisted PD than men [2926/5970 (49.1%) versus 3357/8689 (38.7%)]. In the multivariable analysis, women had a lower risk of transfer to HD [cause-specific hazard ratio {cs-HR} 0.82 {95% confidence interval (CI) 0.77-0.88}], death [cs-HR 0.90 (95% CI 0.85-0.95)], peritonitis [cs-HR 0.82 (95% CI 0.78-0.87)], PD failure [cs-HR 0.86 (95% CI 0.83-0.90)] and a lower chance of undergoing transplant [cs-HR 0.83 (95% CI 0.77-0.90)] than men. There was a direct effect of sex on the risk of transfer to HD [cs-HR 0.82 (95% CI 0.82-0.83)], with an indirect effect of nurse-assisted PD [cs-HR 0.97 (95% CI 0.96-0.99)]. Nurse-assisted PD had no indirect effect on the risk of peritonitis. Our results suggest that compared with men, women have a lower risk of both transfer to HD and peritonitis. Mediation analysis showed that nurse assistance was a potential mediator in the causal pathway between sex and transfer to HD.

Bank systemic risk exposure and office market interconnectedness

We empirically examine how systemic risk in the banking sector leads to correlated risk in office markets of global financial centers. In so doing, we compute an aggregated measure of systemic risk in financial centers as the cumulated expected capital shortfall of local financial institutions. Our identification strategy is based on a double counterfactual approach by comparing normal with financial distress periods as well as office with retail markets. We find that office market interconnectedness arises from systemic risk during financial turmoil periods. Office market performance in a financial center is affected by returns of systemically linked financial center office markets only during a systemic banking crisis. In contrast, there is no evidence of correlated risk during normal times and among the within-city counterfactual retail sector. The decline in office market returns during a banking crisis is larger in financial centers compared to non-financial centers.