Integrating explanation and prediction in computational social science.

Computational social science is more than just large repositories of digital data and the computational methods needed to construct and analyse them. It also represents a convergence of different fields with different ways of thinking about and doing science. The goal of this Perspective is to provide some clarity around how these approaches differ from one another and to propose how they might be productively integrated. Towards this end we make two contributions. The first is a schema for thinking about research activities along two dimensions-the extent to which work is explanatory, focusing on identifying and estimating causal effects, and the degree of consideration given to testing predictions of outcomes-and how these two priorities can complement, rather than compete with, one another. Our second contribution is to advocate that computational social scientists devote more attention to combining prediction and explanation, which we call integrative modelling, and to outline some practical suggestions for realizing this goal.

Digital public health interventions at scale: The impact of social media advertising on beliefs and outcomes related to COVID vaccines.

Public health organizations increasingly use social media advertising campaigns in pursuit of public health goals. In this paper, we evaluate the impact of about $40 million of social media advertisements that were run and experimentally tested on Facebook and Instagram, aimed at increasing COVID-19 vaccination rates in the first year of the vaccine roll-out. The 819 randomized experiments in our sample were run by 174 different public health organizations and collectively reached 2.1 billion individuals in 15 languages. We find that these campaigns are, on average, effective at influencing self-reported beliefs-shifting opinions close to 1% at baseline with a cost per influenced person of about $3.41. Combining this result with an estimate of the relationship between survey outcomes and vaccination rates derived from observational data yields an estimated cost per additional vaccination of about $5.68. There is further evidence that campaigns are especially effective at influencing users' knowledge of how to get vaccines. Our results represent, to the best of our knowledge, the largest set of online public health interventions analyzed to date.

Estimating experienced racial segregation in US cities using large-scale GPS data.

We estimate a measure of segregation, experienced isolation, that captures individuals' exposure to diverse others in the places they visit over the course of their days. Using Global Positioning System (GPS) data collected from smartphones, we measure experienced isolation by race. We find that the isolation individuals experience is substantially lower than standard residential isolation measures would suggest but that experienced isolation and residential isolation are highly correlated across cities. Experienced isolation is lower relative to residential isolation in denser, wealthier, more educated cities with high levels of public transit use and is also negatively correlated with income mobility.

Confidence intervals for policy evaluation in adaptive experiments.

Adaptive experimental designs can dramatically improve efficiency in randomized trials. But with adaptively collected data, common estimators based on sample means and inverse propensity-weighted means can be biased or heavy-tailed. This poses statistical challenges, in particular when the experimenter would like to test hypotheses about parameters that were not targeted by the data-collection mechanism. In this paper, we present a class of test statistics that can handle these challenges. Our approach is to adaptively reweight the terms of an augmented inverse propensity-weighting estimator to control the contribution of each term to the estimator's variance. This scheme reduces overall variance and yields an asymptotically normal test statistic. We validate the accuracy of the resulting estimates and their CIs in numerical experiments and show that our methods compare favorably to existing alternatives in terms of mean squared error, coverage, and CI size.

Federated causal inference in heterogeneous observational data.

We are interested in estimating the effect of a treatment applied to individuals at multiple sites, where data is stored locally for each site. Due to privacy constraints, individual-level data cannot be shared across sites; the sites may also have heterogeneous populations and treatment assignment mechanisms. Motivated by these considerations, we develop federated methods to draw inferences on the average treatment effects of combined data across sites. Our methods first compute summary statistics locally using propensity scores and then aggregate these statistics across sites to obtain point and variance estimators of average treatment effects. We show that these estimators are consistent and asymptotically normal. To achieve these asymptotic properties, we find that the aggregation schemes need to account for the heterogeneity in treatment assignments and in outcomes across sites. We demonstrate the validity of our federated methods through a comparative study of two large medical claims databases.

Recursive partitioning for heterogeneous causal effects.

In this paper we propose methods for estimating heterogeneity in causal effects in experimental and observational studies and for conducting hypothesis tests about the magnitude of differences in treatment effects across subsets of the population. We provide a data-driven approach to partition the data into subpopulations that differ in the magnitude of their treatment effects. The approach enables the construction of valid confidence intervals for treatment effects, even with many covariates relative to the sample size, and without "sparsity" assumptions. We propose an "honest" approach to estimation, whereby one sample is used to construct the partition and another to estimate treatment effects for each subpopulation. Our approach builds on regression tree methods, modified to optimize for goodness of fit in treatment effects and to account for honest estimation. Our model selection criterion anticipates that bias will be eliminated by honest estimation and also accounts for the effect of making additional splits on the variance of treatment effect estimates within each subpopulation. We address the challenge that the "ground truth" for a causal effect is not observed for any individual unit, so that standard approaches to cross-validation must be modified. Through a simulation study, we show that for our preferred method honest estimation results in nominal coverage for 90% confidence intervals, whereas coverage ranges between 74% and 84% for nonhonest approaches. Honest estimation requires estimating the model with a smaller sample size; the cost in terms of mean squared error of treatment effects for our preferred method ranges between 7-22%.

Battling the coronavirus 'infodemic' among social media users in Kenya and Nigeria.

How can we induce social media users to be discerning when sharing information during a pandemic? An experiment on Facebook Messenger with users from Kenya (n = 7,498) and Nigeria (n = 7,794) tested interventions designed to decrease intentions to share COVID-19 misinformation without decreasing intentions to share factual posts. The initial stage of the study incorporated: (1) a factorial design with 40 intervention combinations; and (2) a contextual adaptive design, increasing the probability of assignment to treatments that worked better for previous subjects with similar characteristics. The second stage evaluated the best-performing treatments and a targeted treatment assignment policy estimated from the data. We precisely estimate null effects from warning flags and related article suggestions, tactics used by social media platforms. However, nudges to consider the accuracy of information reduced misinformation sharing relative to control by 4.9% (estimate = -2.3 percentage points, 95% CI = [-4.2, -0.35]). Such low-cost scalable interventions may improve the quality of information circulating online.

Machine-learning-based high-benefit approach versus conventional high-risk approach in blood pressure management.

BACKGROUND: In medicine, clinicians treat individuals under an implicit assumption that high-risk patients would benefit most from the treatment ('high-risk approach'). However, treating individuals with the highest estimated benefit using a novel machine-learning method ('high-benefit approach') may improve population health outcomes. METHODS: This study included 10 672 participants who were randomized to systolic blood pressure (SBP) target of either <120 mmHg (intensive treatment) or <140 mmHg (standard treatment) from two randomized controlled trials (Systolic Blood Pressure Intervention Trial, and Action to Control Cardiovascular Risk in Diabetes Blood Pressure). We applied the machine-learning causal forest to develop a prediction model of individualized treatment effect (ITE) of intensive SBP control on the reduction in cardiovascular outcomes at 3 years. We then compared the performance of high-benefit approach (treating individuals with ITE >0) versus the high-risk approach (treating individuals with SBP ≥130 mmHg). Using transportability formula, we also estimated the effect of these approaches among 14 575 US adults from National Health and Nutrition Examination Surveys (NHANES) 1999-2018. RESULTS: We found that 78.9% of individuals with SBP ≥130 mmHg benefited from the intensive SBP control. The high-benefit approach outperformed the high-risk approach [average treatment effect (95% CI), +9.36 (8.33-10.44) vs +1.65 (0.36-2.84) percentage point; difference between these two approaches, +7.71 (6.79-8.67) percentage points, P-value <0.001]. The results were consistent when we transported the results to the NHANES data. CONCLUSIONS: The machine-learning-based high-benefit approach outperformed the high-risk approach with a larger treatment effect. These findings indicate that the high-benefit approach has the potential to maximize the effectiveness of treatment rather than the conventional high-risk approach, which needs to be validated in future research.

Can personalized digital counseling improve consumer search for modern contraceptive methods?

This paper analyzes a randomized controlled trial of a personalized digital counseling intervention addressing informational constraints and choice architecture, cross-randomized with discounts for long-acting reversible contraceptives (LARCs), such as intrauterine devices (IUDs). The counseling intervention encourages shared decision-making (SDM) using a tablet-based app, which provides a tailored ranking of modern methods to each client according to their elicited needs and preferences. Take-up of LARCs in the status quo regime at full price was 11%, which increased to 28% with discounts. SDM roughly tripled the share of clients adopting a LARC at full price to 35%, and discounts had no incremental impact in this group. Neither intervention affected the take-up of short-acting methods, such as the pill. Consistent with theoretical models of consumer search, SDM clients discussed more methods in depth, which led to higher adoption rates for second- or lower-ranked LARCs. Our findings suggest that low-cost individualized recommendations can potentially be as effective in increasing unfamiliar technology adoption as providing large subsidies.

Uncovering interpretable potential confounders in electronic medical records.

Randomized clinical trials (RCT) are the gold standard for informing treatment decisions. Observational studies are often plagued by selection bias, and expert-selected covariates may insufficiently adjust for confounding. We explore how unstructured clinical text can be used to reduce selection bias and improve medical practice. We develop a framework based on natural language processing to uncover interpretable potential confounders from text. We validate our method by comparing the estimated hazard ratio (HR) with and without the confounders against established RCTs. We apply our method to four cohorts built from localized prostate and lung cancer datasets from the Stanford Cancer Institute and show that our method shifts the HR estimate towards the RCT results. The uncovered terms can also be interpreted by oncologists for clinical insights. We present this proof-of-concept study to enable more credible causal inference using observational data, uncover meaningful insights from clinical text, and inform high-stakes medical decisions.

The Association Between Alpha-1 Adrenergic Receptor Antagonists and In-Hospital Mortality from COVID-19.

Effective therapies for coronavirus disease 2019 (COVID-19) are urgently needed, and preclinical data suggest alpha-1 adrenergic receptor antagonists (α1-AR antagonists) may be effective in reducing mortality related to hyperinflammation independent of etiology. Using a retrospective cohort design with patients in the Department of Veterans Affairs healthcare system, we use doubly robust regression and matching to estimate the association between baseline use of α1-AR antagonists and likelihood of death due to COVID-19 during hospitalization. Having an active prescription for any α1-AR antagonist (tamsulosin, silodosin, prazosin, terazosin, doxazosin, or alfuzosin) at the time of admission had a significant negative association with in-hospital mortality (relative risk reduction 18%; odds ratio 0.73; 95% CI 0.63 to 0.85; p ≤ 0.001) and death within 28 days of admission (relative risk reduction 17%; odds ratio 0.74; 95% CI 0.65 to 0.84; p ≤ 0.001). In a subset of patients on doxazosin specifically, an inhibitor of all three alpha-1 adrenergic receptors, we observed a relative risk reduction for death of 74% (odds ratio 0.23; 95% CI 0.03 to 0.94; p = 0.028) compared to matched controls not on any α1-AR antagonist at the time of admission. These findings suggest that use of α1-AR antagonists may reduce mortality in COVID-19, supporting the need for randomized, placebo-controlled clinical trials in patients with early symptomatic infection.

The Association Between Alpha-1 Adrenergic Receptor Antagonists and In-Hospital Mortality From COVID-19.

Effective therapies for coronavirus disease 2019 (COVID-19) are urgently needed, and pre-clinical data suggest alpha-1 adrenergic receptor antagonists (α1-AR antagonists) may be effective in reducing mortality related to hyperinflammation independent of etiology. Using a retrospective cohort design with patients in the Department of Veterans Affairs healthcare system, we use doubly robust regression and matching to estimate the association between baseline use of α1-AR antagonists and likelihood of death due to COVID-19 during hospitalization. Having an active prescription for any α1-AR antagonist (tamsulosin, silodosin, prazosin, terazosin, doxazosin, or alfuzosin) at the time of admission had a significant negative association with in-hospital mortality (relative risk reduction 18%; odds ratio 0.73; 95% CI 0.63-0.85; p ≤ 0.001) and death within 28 days of admission (relative risk reduction 17%; odds ratio 0.74; 95% CI 0.65-0.84; p ≤ 0.001). In a subset of patients on doxazosin specifically, an inhibitor of all three alpha-1 adrenergic receptors, we observed a relative risk reduction for death of 74% (odds ratio 0.23; 95% CI 0.03-0.94; p = 0.028) compared to matched controls not on any α1-AR antagonist at the time of admission. These findings suggest that use of α1-AR antagonists may reduce mortality in COVID-19, supporting the need for randomized, placebo-controlled clinical trials in patients with early symptomatic infection.

Association of α1-Blocker Receipt With 30-Day Mortality and Risk of Intensive Care Unit Admission Among Adults Hospitalized With Influenza or Pneumonia in Denmark.

Importance: Alpha 1-adrenergic receptor blocking agents (α1-blockers) have been reported to have protective benefits against hyperinflammation and cytokine storm syndrome, conditions that are associated with mortality in patients with coronavirus disease 2019 and other severe respiratory tract infections. However, studies of the association of α1-blockers with outcomes among human participants with respiratory tract infections are scarce. Objective: To examine the association between the receipt of α1-blockers and outcomes among adult patients hospitalized with influenza or pneumonia. Design, Setting, and Participants: This population-based cohort study used data from Danish national registries to identify individuals 40 years and older who were hospitalized with influenza or pneumonia between January 1, 2005, and November 30, 2018, with follow-up through December 31, 2018. In the main analyses, patients currently receiving α1-blockers were compared with those not receiving α1-blockers (defined as patients with no prescription for an α1-blocker filled within 365 days before the index date) and those currently receiving 5α-reductase inhibitors. Propensity scores were used to address confounding factors and to compute weighted risks, absolute risk differences, and risk ratios. Data were analyzed from April 21 to December 21, 2020. Exposures: Current receipt of α1-blockers compared with nonreceipt of α1-blockers and with current receipt of 5α-reductase inhibitors. Main Outcomes and Measures: Death within 30 days of hospital admission and risk of intensive care unit (ICU) admission. Results: A total of 528 467 adult patients (median age, 75.0 years; interquartile range, 64.4-83.6 years; 273 005 men [51.7%]) were hospitalized with influenza or pneumonia in Denmark between 2005 and 2018. Of those, 21 772 patients (4.1%) were currently receiving α1-blockers compared with a population of 22 117 patients not receiving α1-blockers who were weighted to the propensity score distribution of those receiving α1-blockers. In the propensity score-weighted analyses, patients receiving α1-blockers had lower 30-day mortality (15.9%) compared with patients not receiving α1-blockers (18.5%), with a corresponding risk difference of -2.7% (95% CI, -3.2% to -2.2%) and a risk ratio (RR) of 0.85 (95% CI, 0.83-0.88). The risk of ICU admission was 7.3% among patients receiving α1-blockers and 7.7% among those not receiving α1-blockers (risk difference, -0.4% [95% CI, -0.8% to 0%]; RR, 0.95 [95% CI, 0.90-1.00]). A comparison between 18 280 male patients currently receiving α1-blockers and 18 228 propensity score-weighted male patients currently receiving 5α-reductase inhibitors indicated that those receiving α1-blockers had lower 30-day mortality (risk difference, -2.0% [95% CI, -3.4% to -0.6%]; RR, 0.89 [95% CI, 0.82-0.96]) and a similar risk of ICU admission (risk difference, -0.3% [95% CI, -1.4% to 0.7%]; RR, 0.96 [95% CI, 0.83-1.10]). Conclusions and Relevance: This cohort study's findings suggest that the receipt of α1-blockers is associated with protective benefits among adult patients hospitalized with influenza or pneumonia.

Ten Rules for Conducting Retrospective Pharmacoepidemiological Analyses: Example COVID-19 Study.

Since the beginning of the COVID-19 pandemic, pharmaceutical treatment hypotheses have abounded, each requiring careful evaluation. A randomized controlled trial generally provides the most credible evaluation of a treatment, but the efficiency and effectiveness of the trial depend on the existing evidence supporting the treatment. The researcher must therefore compile a body of evidence justifying the use of time and resources to further investigate a treatment hypothesis in a trial. An observational study can provide this evidence, but the lack of randomized exposure and the researcher's inability to control treatment administration and data collection introduce significant challenges. A proper analysis of observational health care data thus requires contributions from experts in a diverse set of topics ranging from epidemiology and causal analysis to relevant medical specialties and data sources. Here we summarize these contributions as 10 rules that serve as an end-to-end introduction to retrospective pharmacoepidemiological analyses of observational health care data using a running example of a hypothetical COVID-19 study. A detailed supplement presents a practical how-to guide for following each rule. When carefully designed and properly executed, a retrospective pharmacoepidemiological analysis framed around these rules will inform the decisions of whether and how to investigate a treatment hypothesis in a randomized controlled trial. This work has important implications for any future pandemic by prescribing what we can and should do while the world waits for global vaccine distribution.

Falling living standards during the COVID-19 crisis: Quantitative evidence from nine developing countries.

Despite numerous journalistic accounts, systematic quantitative evidence on economic conditions during the ongoing COVID-19 pandemic remains scarce for most low- and middle-income countries, partly due to limitations of official economic statistics in environments with large informal sectors and subsistence agriculture. We assemble evidence from over 30,000 respondents in 16 original household surveys from nine countries in Africa (Burkina Faso, Ghana, Kenya, Rwanda, Sierra Leone), Asia (Bangladesh, Nepal, Philippines), and Latin America (Colombia). We document declines in employment and income in all settings beginning March 2020. The share of households experiencing an income drop ranges from 8 to 87% (median, 68%). Household coping strategies and government assistance were insufficient to sustain precrisis living standards, resulting in widespread food insecurity and dire economic conditions even 3 months into the crisis. We discuss promising policy responses and speculate about the risk of persistent adverse effects, especially among children and other vulnerable groups.

Alpha-1 adrenergic receptor antagonists to prevent hyperinflammation and death from lower respiratory tract infection.

In severe viral pneumonia, including Coronavirus disease 2019 (COVID-19), the viral replication phase is often followed by hyperinflammation, which can lead to acute respiratory distress syndrome, multi-organ failure, and death. We previously demonstrated that alpha-1 adrenergic receptor ($\alpha_1$-AR) antagonists can prevent hyperinflammation and death in mice. Here, we conducted retrospective analyses in two cohorts of patients with acute respiratory distress (ARD, n=18,547) and three cohorts with pneumonia (n=400,907). Federated across two ARD cohorts, we find that patients exposed to $\alpha_1$-AR antagonists, as compared to unexposed patients, had a 34% relative risk reduction for mechanical ventilation and death (OR=0.70, p=0.021). We replicated these methods on three pneumonia cohorts, all with similar effects on both outcomes. All results were robust to sensitivity analyses. These results highlight the urgent need for prospective trials testing whether prophylactic use of $\alpha_1$-AR antagonists ameliorates lower respiratory tract infection-associated hyperinflammation and death, as observed in COVID-19.

Alpha-1 adrenergic receptor antagonists to prevent hyperinflammation and death from lower respiratory tract infection.

In severe viral pneumonia, including Coronavirus disease 2019 (COVID-19), the viral replication phase is often followed by hyperinflammation, which can lead to acute respiratory distress syndrome, multi-organ failure, and death. We previously demonstrated that alpha-1 adrenergic receptor (⍺1-AR) antagonists can prevent hyperinflammation and death in mice. Here, we conducted retrospective analyses in two cohorts of patients with acute respiratory distress (ARD, n = 18,547) and three cohorts with pneumonia (n = 400,907). Federated across two ARD cohorts, we find that patients exposed to ⍺1-AR antagonists, as compared to unexposed patients, had a 34% relative risk reduction for mechanical ventilation and death (OR = 0.70, p = 0.021). We replicated these methods on three pneumonia cohorts, all with similar effects on both outcomes. All results were robust to sensitivity analyses. These results highlight the urgent need for prospective trials testing whether prophylactic use of ⍺1-AR antagonists ameliorates lower respiratory tract infection-associated hyperinflammation and death, as observed in COVID-19.

Beyond prediction: Using big data for policy problems.

Machine-learning prediction methods have been extremely productive in applications ranging from medicine to allocating fire and health inspectors in cities. However, there are a number of gaps between making a prediction and making a decision, and underlying assumptions need to be understood in order to optimize data-driven decision-making.