Concept: Propensity score
Previous research suggests a link between the quality of teacher-student relationships and the students' behavioral outcomes; however, the observational nature of past studies makes it difficult to attribute a causal role to the quality of these relationships. In the current study, therefore, we used a propensity score analysis approach to evaluate whether students who were matched on their propensity to experience a given level of relationship quality but differed on their actual relationship quality diverged on their concurrent and subsequent problem and prosocial behavior. Student/self, teacher, and parent- (only waves 1-3) reported data from 8 waves of the Zurich Project on the Social Development of Children and Youths (z-proso), a longitudinal study of Swiss youth among a culturally diverse sample of 7- to 15-year-olds were utilized. The initial sample included 1483 (49.4 % female) students for whom information relevant for this study was available. The sample represented families from around 80 different countries, from across all the continents; with approximately 42 % of the female primary caregivers having been born in Switzerland. Following successful matching, we found that students who reported better relationships with their teachers and whose teachers reported better relationships with them evidenced fewer problem behaviors concurrently and up to 4 years later. There was also evidence for an analogous effect in predicting prosocial behavior. The implications of these findings are discussed in relation to prevention and intervention practices.
When estimating the average effect of a binary treatment (or exposure) on an outcome, methods that incorporate propensity scores, the G-formula, or targeted maximum likelihood estimation (TMLE) are preferred over naïve regression approaches, which are biased under misspecification of a parametric outcome model. In contrast propensity score methods require the correct specification of an exposure model. Double-robust methods only require correct specification of either the outcome or the exposure model. Targeted maximum likelihood estimation is a semiparametric double-robust method that improves the chances of correct model specification by allowing for flexible estimation using (nonparametric) machine-learning methods. It therefore requires weaker assumptions than its competitors. We provide a step-by-step guided implementation of TMLE and illustrate it in a realistic scenario based on cancer epidemiology where assumptions about correct model specification and positivity (ie, when a study participant had 0 probability of receiving the treatment) are nearly violated. This article provides a concise and reproducible educational introduction to TMLE for a binary outcome and exposure. The reader should gain sufficient understanding of TMLE from this introductory tutorial to be able to apply the method in practice. Extensive R-code is provided in easy-to-read boxes throughout the article for replicability. Stata users will find a testing implementation of TMLE and additional material in the Appendix S1 and at the following GitHub repository: https://github.com/migariane/SIM-TMLE-tutorial.
Objectives To study the association between maternal use of antidepressants during pregnancy and autism spectrum disorder (ASD) in offspring.Design Observational prospective cohort study with regression methods, propensity score matching, sibling controls, and negative control comparison.Setting Stockholm County, Sweden.Participants 254 610 individuals aged 4-17, including 5378 with autism, living in Stockholm County in 2001-11 who were born to mothers who did not take antidepressants and did not have any psychiatric disorder, mothers who took antidepressants during pregnancy, or mothers with psychiatric disorders who did not take antidepressants during pregnancy. Maternal antidepressant use was recorded during first antenatal interview or determined from prescription records.Main outcome measure Offspring diagnosis of autism spectrum disorder, with and without intellectual disability.Results Of the 3342 children exposed to antidepressants during pregnancy, 4.1% (n=136) had a diagnosis of autism compared with a 2.9% prevalence (n=353) in 12 325 children not exposed to antidepressants whose mothers had a history of a psychiatric disorder (adjusted odds ratio 1.45, 95% confidence interval 1.13 to 1.85). Propensity score analysis led to similar results. The results of a sibling control analysis were in the same direction, although with wider confidence intervals. In a negative control comparison, there was no evidence of any increased risk of autism in children whose fathers were prescribed antidepressants during the mothers' pregnancy (1.13, 0.68 to 1.88). In all analyses, the risk increase concerned only autism without intellectual disability.Conclusions The association between antidepressant use during pregnancy and autism, particularly autism without intellectual disability, might not solely be a byproduct of confounding. Study of the potential underlying biological mechanisms could help the understanding of modifiable mechanisms in the aetiology of autism. Importantly, the absolute risk of autism was small, and, hypothetically, if no pregnant women took antidepressants, the number of cases that could potentially be prevented would be small.
Clinical guidelines recommend early discharge of patients with low-risk pulmonary embolism (LRPE). This study measured the overall impact of early discharge of LRPE patients on clinical outcomes and costs in the Veterans Health Administration population. Adult patients with ≥1 inpatient diagnosis for pulmonary embolism (PE) (index date) between 10/2011-06/2015, continuous enrollment for ≥12 months pre- and 3 months post-index date were included. PE risk stratification was performed using the simplified Pulmonary Embolism Stratification Index. Propensity score matching (PSM) was used to compare 90-day adverse PE events (APEs) [recurrent venous thromboembolism, major bleed and death], hospital-acquired complications (HACs), healthcare utilization, and costs among short (≤2 days) versus long length of stay (LOS). Net clinical benefit was defined as 1 minus the combined rate of APE and HAC. Among 6,746 PE patients, 95.4% were men, 22.0% were African American, and 1,918 had LRPE. Among LRPE patients, only 688 had a short LOS. After 1:1 PSM, there were no differences in APE, but short LOS had fewer HAC (1.5% vs 13.3%, 95% CI: 3.77-19.94) and bacterial pneumonias (5.9% vs 11.7%, 95% CI: 1.24-3.23), resulting in better net clinical benefit (86.9% vs 78.3%, 95% CI: 0.84-0.96). Among long LOS patients, HACs (52) exceeded APEs (14 recurrent DVT, 5 bleeds). Short LOS incurred lower inpatient ($2,164 vs $5,100, 95% CI: $646.8-$5225.0) and total costs ($9,056 vs $12,544, 95% CI: $636.6-$6337.7). LRPE patients with short LOS had better net clinical outcomes at lower costs than matched LRPE patients with long LOS.
There is mixed evidence from correlational studies that breastfeeding impacts children’s development. Propensity score matching with large samples can be an effective tool to remove potential bias from observed confounders in correlational studies. The aim of this study was to investigate the impact of breastfeeding on children’s cognitive and noncognitive development at 3 and 5 years of age.
OBJECTIVE:: In a large nationwide administrative database of hospitalized patients, we investigated postoperative outcomes after laparoscopic or open distal gastrectomy in Japan. BACKGROUND:: The benefits of laparoscopic gastrectomy, such as decreased length of stay and morbidity, have typically been evaluated only with limited data on the basis of small samples. METHODS:: Using the Japanese Diagnosis Procedure Combination Database, we identified 9388 patients who were preoperatively diagnosed with stage I and II gastric cancer and underwent laparoscopic (n = 3937) or open (n = 5451) distal gastrectomy between July and December 2010. One-to-one propensity score matching was performed to compare in-hospital mortality, postoperative complication rates, length of stay, total costs, and 30-day readmission rates between the 2 groups. RESULTS:: Patients with younger age, lower comorbidity index, or stage I cancer were more likely to receive laparoscopic gastrectomy. In the propensity-matched analysis with 2473 pairs, the laparoscopic gastrectomy group in comparison with the open gastrectomy group showed a slight reduction in median postoperative length of stay (13 days vs 15 days, P < 0.001) but a slight increase in median total costs (US $21,510 vs $21,024, P = 0.002). There were no significant differences in in-hospital mortality (0.36% vs 0.28%, P = 0.80), overall postoperative complications (12.9% vs 12.6%, P = 0.73), or 30-day readmission rates (3.2% vs 3.2%, P = 0.94). CONCLUSIONS:: In this large nationwide cohort of patients with early-stage gastric cancer, laparoscopic gastrectomy was associated with a statistically significant but slight reduction in postoperative length of stay, but no differences between laparoscopic gastrectomy and open gastrectomy were detected in terms of early mortality and morbidity.
PURPOSE: Heart failure (CHF) guidelines recommend mineralocorticoid receptor antagonists for all symptomatic patients treated with a combination of ACE inhibitors/angiotensin receptor blockers (ARBs) and beta-blockers. As opposed to both eplerenone trials, patients in RALES (spironolactone) received almost no beta-blockers. Since pharmacological properties differ between eplerenone and spironolactone, the prognostic benefit of spironolactone added to this baseline combination therapy needs clarification. METHODS: We included 4,832 CHF patients with chronic systolic dysfunction from the Norwegian Heart Failure Registry and the heart failure outpatients' clinic of the University of Heidelberg. Propensity scores for spironolactone receipt were calculated for each patient and used for matching to patients without spironolactone. RESULTS: During a total follow-up of 17,869 patient-years, 881 patients (27.0 %) died in the non-spironolactone group and 445 (28.4 %) in the spironolactone group. Spironolactone was not associated with improved survival, neither in the complete sample (HR 0.82; 95 % CI 0.64-1.07; HR 1.03; 95 % CI 0.88-1.20; multivariate and propensity score adjusted respectively), nor in the propensity-matched cohort (HR 0.98; 95 % CI 0.82-1.18). CONCLUSION: In CHF outpatients we were unable to observe an association between the use of spironolactone and improved survival when administered in addition to a combination of ACE/ARB and beta-blockers.
The purpose of this study was to examine the association between oral statin use and the progression of open angle glaucoma.
The objective of this retrospective study was to assess perioperative outcomes, overall survival and freedom from recurrence after induction chemotherapy followed by extrapleural pneumonectomy (EPP) or pleurectomy/decortication (P/D) in patients with mesothelioma in a propensity score matched analysis.
In this retrospective cohort study, we compared the survival of patients detected by screening with those detected based on symptoms, according to their tumor stages. After propensity score matching, 2,130 patients with papillary or follicular thyroid cancer, identified by screening detection (SD) and clinical detection (CD), were included. We compared the survival rates of patients identified by SD and CD in the early and advanced stages of thyroid cancer. Cox proportional hazard models were used to compare the hazard ratios (HRs) for mortality between the two groups. Of the 1,065 patients in each group, 12 (1.1%) died in the SD group, compared to 44 (4.1%) in the CD group, during an average 9.4 years (p<0.001). For early stage, there was no significant difference in all-cause and thyroid cancer-specific mortality between the two groups (p = 0.08, p = 0.0502). However, for advanced stage, the survival rates in the SD group were significantly higher than in the CD group (p<0.001, p = 0.004). Moreover, after adjusting for covariates, the HRs of all-cause mortality of the SD group was significantly lower than that of the CD group for the advanced stage patients (HRs: 0.37 [95% CIs: 0.17-0.80]), while no significant difference was observed in the early stage. While screening for thyroid cancer was not beneficial for early stage patients, our findings suggest that detection via screening is associated with better survival for patients with advanced stage cancer. However, the effects of selection bias and lead time bias could not be entirely excluded.