Generalized Propensity Score Approach Research Articles

Addressing endogeneity between hazardous actions and motorcyclist injury severity by integrating generalized propensity score approach and instrumental variable model

Motorcyclist hazardous actions (e.g., particularly speed too fast or failing to stop in assured clear distance (ACD)) are commonly identified as risk factors that significantly impact the motorcyclist injury severity. However, endogenous effects resulting from motorcyclist hazardous actions have seldom been considered, which may cause the biased estimates. Specifically, two important sources of endogeneities (i.e., endogeneity arising from observed confounding factors and endogeneity caused by unobserved confounders) tend to yield a biased relationship between hazardous actions and motorcyclist injury severity. To jointly account for two sources of endogeneities and provide more robust estimates, the study tries to assess the effects of speed-too-fast and failing to stop in ACD on motorcyclist injury severity via a hybrid method by integrating the generalized propensity score approach with instrumental variable model. Specifically, we adopt a generalized propensity score matching method to reduce the endogeneity bias arising from observed confounders. Furthermore, the matched data are used to develop an instrumental variable model with random parameters to handle the endogeneity resulting from unobserved confounders and unobserved heterogeneity, which consists of random parameters binary logit models modelling the motorcyclist hazardous actions in the first stage and a random parameters logit model with heterogeneity in means modelling the motorcyclist injury severity in the second stage. The proposed approach is estimated based on Michigan motorcycle crash data from 2015 to 2018. Results suggest that alcohol use leads motorcyclists to engage in speed-too-fast, while alcohol use and signal control cause motorcyclists to be involved in failing to stop in ACD. Middle-aged and elderly motorcyclists, alcohol use, speed too fast, speed limit ≥50 mph, wet surface, and head-on/angle crashes significantly increase the injury severity of motorcyclists. Moreover, failing to stop in ACD produces a random parameter with heterogeneity in means, while intersection increases the mean effects of failing to stop in ACD on motorcyclist minor injury. These findings further provide insights for a better understanding of hazardous actions and motorcyclist injury severity via the impact analysis of various explanatory variables.

Subcutaneous Adipose Tissue Is Associated with Acute Kidney Injury after Abdominal Trauma Based on the Generalized Propensity Score Approach: A Retrospective Cohort Study

Introduction: Obesity is associated with an increased risk of acute kidney injury (AKI) after trauma. However, the associations between different adipose tissue depots and AKI remain unknown. Our study aimed to quantify the effect of abdominal adiposity on AKI in trauma patients. Methods: We performed a retrospective cohort study of abdominal trauma patients who were admitted to our hospital from January 2010 to March 2020. Abdominal visceral adipose tissue (VAT) and subcutaneous adipose tissue (SAT) were measured at the level of the third lumbar vertebra using computed tomography. Causal modeling based on the generalized propensity score was used to quantify the effects of body mass index (BMI), VAT, and SAT on AKI. Results: Among 324 abdominal trauma patients, 67 (20.68%) patients developed AKI. Patients with AKI had higher BMI (22.46 kg/m² vs. 22.04 kg/m², p = 0.014), higher SAT areas (89.06 cm² vs. 83.39 cm², p = 0.151), and higher VAT areas (140.02 cm² vs. 91.48 cm², p = 0.001) than those without AKI. By using causal modeling, we found that the risk of developing AKI increased by 8.3% (p = 0.001) and 4.8% (p = 0.022) with one unit increase in BMI (per 1 kg/m²) and ten units increase in SAT (per 10 cm²), respectively. However, VAT did not show a significant association with AKI (p = 0.327). Conclusion: SAT, but not VAT, increased the risk of AKI among abdominal trauma patients. Measurement of SAT might help identify patients at higher risk of AKI.

Generalized propensity score approach to causal inference with spatial interference.

Many spatial phenomena exhibit interference, where exposures at one location may affect the response at other locations. Because interference violates the stable unit treatment value assumption, standard methods for causal inference do not apply. We propose a new causal framework to recover direct and spill-over effects in the presence of spatial interference, taking into account that exposures at nearby locations are more influential than exposures at locations further apart. Under the no unmeasured confounding assumption, we show that a generalized propensity score is sufficient to remove all measured confounding. To reduce dimensionality issues, we propose a Bayesian spline-based regression model accounting for a sufficient set of variables for the generalized propensity score. A simulation study demonstrates the accuracy and coverage properties. We apply the method to estimate the causal effect of wildland fires on air pollution in the Western United States over2005-2018.

The Effect of Community Socioeconomic Context on High School Attendance in China: A Generalized Propensity Score Approach

Many developing countries have experienced increasing spatial inequality, but little is known about the effect of community disadvantages on educational attainment in these societies. Using data from the China Family Panel Studies (2010–2016), I examine the effect of community socioeconomic status (SES) on the transition into high school in urban and rural China, and I explore several mechanisms explaining the community effects. I adopt the generalized propensity score method to estimate the potential probability of high school entrance at different levels of community SES. Results show that community SES is positively associated with high school attendance in both urban and rural China, and the relationship is stronger in more disadvantaged communities in both contexts. In urban areas, the effect of community SES is partly attributable to collective socialization and children’s academic performance. In rural areas, spatial accessibility to high schools and children’s academic performance are the salient mechanisms.

A hybrid generalized propensity score approach for observational studies

The overlap assumption (OA) of the generalized propensity score (GPS) is often violated when the number of treatments increases (Lin et al., 2019), especially when the GPS is estimated using machine learning (ML) algorithms. Although the ML based GPS (ML-GPS) shows better performance than the multinomial logistic regression (MLR) based GPS (MLR-GPS), the ML-GPS frequently suffers from the OA violation, which causes difficulty in estimating average treatment effects (ATE) using weighted estimators. Thus, we propose a hybrid GPS that is easy to implement in practice to overcome the OA violation. The hybrid GPS simply combines the strengths of the ML-GPS and MLR-GPS. We conduct a Monte Carlo simulation to compare MLR-GPS with several hybrid GPS for estimating ATEs in terms of bias and mean squared error (MSE). Results show that, overall, the hybrid GPS performs better than the MLR-GPS.

Household welfare implications of sustainable land management practices among smallholder farmers in Ghana

This study employs farm household data to examine the impact of adoption of sustainable land management practices on consumption and poverty outcomes using multivalued treatment effects approaches. The generalized propensity score approach was employed to account for selectivity bias due to observed characteristics among households. Using a doubly robust treatment effects estimator, we found that increasing intensity of adoption improved per capita consumption expenditure, reduced poverty headcount and poverty gap among farm households. The results of the multivalued treatment effects approach revealed significant differences between adopters of different adoption intensities, a finding that is not possible with binary treatment effect approaches. The findings also revealed that the average treatment effect of moving from low intensity to high intensity adoption levels differed across quantiles of per capita consumption. We also use a dose-response function to demonstrate that the treatment effect of intensity of adoption on per capita consumption and poverty outcomes is nonlinear, with optimal adoption level occurring between 60–70 % of adoption intensity dose. It is therefore recommended that initial government support in sustainable land management is necessary to sustain and upscale individual household efforts towards investment in sustainable land management.

Cognitive ability and teen smoking.

We use data on 11-15-year-old boys in the West Bank to study the empirical link between cognitive ability and health behavior, specifically (teen) smoking. Adjusting for both age in months and grade level allows us to effectively shut down any simultaneous effect of maturation and schooling on cognitive ability and smoking. We find that those at the lower end of the cognitive ability distribution are more than twice as likely to smoke than those at the upper end (approximately 25 versus 10%) also after adjusting for parental background and peer composition in a generalized propensity score approach. Further, we find that the cognitive ability-smoking gradient is fairly flat at the lower end of the cognitive ability distribution and steep at the upper end.

Estimating short term impact of fine airborne particles on mortality using a semiparametric Generalized Propensity Score approach

Estimating short term impact of fine airborne particles on mortality using a semiparametric Generalized Propensity Score approach

Measuring the effects of advertising on green industry sales: a generalized propensity score approach

ABSTRACTThis article estimates the effects of advertising expenditures on annual gross sales of green industry firms using a quasi-experimental framework. In order to account for potential selection bias, a generalized propensity score and a dose-response function are used to estimate advertising treatment effects. The method used allows us to investigate the relationship between the dose (advertising expenditures) and the response (firm sales). We use data from the National Green Industry Surveys of 2009 and 2014 to conduct the analysis. To further investigate potential heterogeneous advertising effects of the size of the firms, we separate the sample into small firms and large firms, according to their annual gross sales. The results indicate that the magnitude and shape of the response function depend on the size of the firm. For small firms, increasing advertising spending yields to higher sales within a range of advertising spending. Beyond this range, advertising spending increases do not impact sales any more. Thus, small firms’ management should carefully monitor advertising input. For large firms, on the other hand, the current evidence does not support a positive relationship between advertising spending and sales since the marginal treatment effect is insignificant almost over the entire range of adverting spending.

Circulating Cholesterol Levels May Link to the Factors Influencing Parkinson's Risk.

A growing literature suggests that circulating cholesterol levels have been associated with Parkinson's disease (PD). In this study, we investigated a possible causal basis for the cholesterol-PD link. Fasting plasma cholesterol levels were obtained from 91 PD and 70 age- and gender-matched controls from an NINDS PD Biomarkers Program cohort at the Pennsylvania State University College of Medicine. Based on the literature, genetic polymorphisms in selected cholesterol management genes (APOE, LDLR, LRP1, and LRPAP1) were chosen as confounding variables because they may influence both cholesterol levels and PD risk. First, the marginal structure model was applied, where the associations of total- and LDL-cholesterol levels with genetic polymorphisms, statin usage, and smoking history were estimated using linear regression. Then, potential causal influences of total- and LDL-cholesterol on PD occurrence were investigated using a generalized propensity score approach in the second step. Both statins (p < 0.001) and LRP1 (p < 0.03) influenced total- and LDL-cholesterol levels. There also was a trend for APOE to affect total- and LDL-cholesterol (p = 0.08 for both), and for LRPAR1 to affect LDL-cholesterol (p = 0.05). Conversely, LDLR did not influence plasma cholesterol levels (p > 0.19). Based on propensity score methods, lower total- and LDL-cholesterol were significantly linked to PD (p < 0.001 and p = 0.04, respectively). The current study suggests that circulating total- and LDL-cholesterol levels potentially may be linked to the factor(s) influencing PD risk. Further studies to validate these results would impact our understanding of the role of cholesterol as a risk factor in PD, and its relationship to recent public health controversies.

How does the presence of HOV lanes affect plug-in electric vehicle adoption in California? A generalized propensity score approach

Policymakers have sought to spur consumer adoption of advanced clean vehicles by granting them single-occupancy access to high-occupancy vehicle (HOV) lanes. We offer the first evaluation of these policies that accommodates geographic variability in the magnitude of this policy's treatment effect. Focusing on the outcome of plug-in electric vehicle (PEV) adoption in California, we employ a generalized propensity score approach that allows for continuous, rather than binary, treatment effects. We estimate a state-wide dose-response curve to show that access to 6, 20, and 100 miles of nearby HOV lanes leads to 1, 3, and 10 additional PEV registrations in a census tract. The lower end of our 95% confidence interval implies that at least one quarter of California PEV registrations during 2010–2013 were a result of the HOV lane policy. We identify geographically-specific marginal policy effects that are smaller in Los Angeles, but relatively larger in San Diego and Sacramento.

Assessing the dose-response relationship between maternal use of inhaled corticosteroids therapy and birth weight: a generalized propensity score approach

Assessing the dose-response relationship between maternal use of inhaled corticosteroids therapy and birth weight: a generalized propensity score approach

Organic certification, agro-ecological practices and return on investment: Evidence from pineapple producers in Ghana

The recent empirical literature on economic sustainability of certified export crops shows that certification standards that enhance yields are important for improving farm revenues and household welfare. However, limited evidence exists on the impact of organic certification on the adoption of agro-ecological practices. In this study, we use unique farm-level data from pineapple producers in Ghana to examine the impact of organic certification on the use of agro-ecological practices such as organic fertilizers, organic pest and weed control, crop rotation, and soil and water conservation, as well as how using these measures affect farm outcomes such as return on investment. Our empirical results reveal that organic certification increases agro-ecological practice use, although from a very low starting point. Using a generalized propensity score approach, we show that there is a positive, but nonlinear relationship between the intensity of agro-ecological practice use and return on investment.

Duration of breastfeeding and childhood obesity: a generalized propensity score approach.

To estimate the effect of breastfeeding duration on childhood obesity. The Child Development Supplement (CDS) of the Panel Study of Income Dynamics (PSID). The PSID provides extensive data on the income and well-being of a representative sample of U.S. families from 1968 to the present. The CDS collects information on the children in PSID families ranging from cognitive, behavioral, and health status to their family and neighborhood environment. The first two waves of the CDS were conducted in 1997 and 2002, respectively. The data provide information on 3,271 children and their mothers. We use the generalized propensity score to adjust for confounding based on continuous treatment, and the general additive model to analyze the adjusted association between treatment and the outcome conditional on the propensity score. The main outcome is the body mass index (BMI) directly assessed during the in-person interview in 2002. Covariates include family, maternal, and child characteristics, many of which were measured in the year the child was born. After using propensity scores to adjust for confounding, the relationship between breastfeeding duration and childhood BMI is trivially small across a range of model specifications, and none of them is statistically significant except the unadjusted model. The causal link between duration of breastfeeding and childhood obesity has not been established. Any recommendation of promoting breastfeeding to reduce childhood obesity is premature.

Effectiveness of Ribavirin and Corticosteroids for Severe Acute Respiratory Syndrome

Objective Ribavirin and corticosteroids were used widely as front-line treatments for severe acute respiratory syndrome; however, previous evaluations were inconclusive. We assessed the effectiveness of ribavirin and corticosteroids as the initial treatment for severe acute respiratory syndrome using propensity score analysis. Methods We analyzed data on 1755 patients in Hong Kong and 191 patients in Toronto with severe acute respiratory syndrome using a generalized propensity score approach. Results The adjusted excess case fatality ratios of patients with severe acute respiratory syndrome receiving the combined therapy of ribavirin and corticosteroids within 2 days of admission, compared with those receiving neither treatment within 2 days of admission, were 3.8% (95% confidence interval, −1.5 to 9.2) in Hong Kong and 2.1% (95% confidence interval, −44.3 to 48.5) in Toronto. Conclusions Our results add strength to the hypothesis that the combination of ribavirin and corticosteroids has no therapeutic benefit when given early during severe acute respiratory syndrome infection. Further studies may investigate the effects of these treatments later in disease course.