The popularity of online surveys has increased the prominence of using weights that capture units' probabilities of inclusion for claims of representativeness. Yet, much uncertainty remains regarding how these weights should be employed in analysis of survey experiments: Should they be used or ignored? If they are used, which estimators are preferred? We offer practical advice, rooted in the Neyman-Rubin model, for researchers producing and working with survey experimental data. We examine simple, efficient estimators for analyzing these data, and give formulae for their biases and variances. We provide simulations that examine these estimators as well as real examples from experiments administered online through YouGov. We find that for examining the existence of population treatment effects using highquality, broadly representative samples recruited by top online survey firms, sample quantities, which do not rely on weights, are often sufficient. We found that Sample Average Treatment Effect (SATE) estimates did not appear to differ substantially from their weighted counterparts, and they avoided the substantial loss of statistical power that accompanies weighting. When precise estimates of Population Average Treatment Effects (PATE) are essential, we analytically show post-stratifying on survey weights and/or covariates highly correlated with the outcome to be a conservative choice. While we show these substantial gains in simulations, we find limited evidence of them in practice.

Background: Unplanned hospital readmissions are a major source of morbidity among dialysis patients, in whom the risk of hospital readmission is exceptionally high. The contribution of dialysis facility staffing to hospital readmission has been largely overlooked. Methods: Using annual data of dialysis patients from the United States Renal Data System from 2010 to 2013, we assessed dialysis facilities with a significantly worse (SW) and facilities with a nonsignificant (NS) standardized readmission ratio (SRR). SRR estimates were risk adjusted for patient factors, past year comorbidities, and index hospitalization characteristics. Facility staffing variables were compared between 2 exposure groups: facilities with SW and NS SRRs. Four measures of staffing, including patient-to-staffing ratio, were compared between SW and matched NS facilities. Results: About 136,000–148,000 dialysis patients with 269,000–319,000 index hospital discharges were used to identify facilities with SW and facilities with NS SRR annually. Approximately 3–4% of facilities were identified as having SW SRR among > 5,000 facilities annually. The percent of nurses-to-total staff was significantly lower in 2010 for SW facilities than in matched NS facilities (42.5 vs. 45.6%, p = 0.012), but this disparity was attenuated by 2013 (44.8 vs. 44.7%, p = 0.949). There was a higher patient-to-nurse ratio for SW facilities than for NS facilities (mean 16.4 vs. 15.2, p = 0.038) in 2010 as well, and the disparity was reduced by 2013. The trends were similar for patient-to-total staff and patient-to-registered nurse, but not statistically significant. Conclusions: This study found that dialysis facilities with SW 30-day readmission rates had lower proportions of nurses-to-total staff and higher patient-to-nurse ratios, but this disparity improved in recent years. Additional research is warranted focusing on how evidence-based staffing at dialysis facilities can contribute to reduction of hospital readmission, and this knowledge is needed to inform clinical practice guidelines and policy decisions regarding optimal dialysis patient staffing.

No abstract

Summary One of the most significant barriers to medication treatment is patients’ non-adherence to a prescribed medication regimen. The extent of the impact of poor adherence on resulting health measures is often unknown, and typical analyses ignore the time-varying nature of adherence. This article develops a modeling framework for longitudinally recorded health measures modeled as a function of time-varying medication adherence. Our framework, which relies on normal Bayesian dynamic linear models (DLMs), accounts for time-varying covariates such as adherence and non-dynamic covariates such as baseline health characteristics. Standard inferential procedures for DLMs are inefficient when faced with infrequent and irregularly recorded response data. We develop an approach that relies on factoring the posterior density into a product of two terms: a marginal posterior density for the non-dynamic parameters, and a multivariate normal posterior density of the dynamic parameters conditional on the non-dynamic ones. This factorization leads to a two-stage process for inference in which the non-dynamic parameters can be inferred separately from the time-varying parameters. We demonstrate the application of this model to the time-varying effect of antihypertensive medication on blood pressure levels for a cohort of patients diagnosed with hypertension. Our model results are compared to ones in which adherence is incorporated through non-dynamic summaries.

No abstract