Impact Measurement based on Repeated Randomized Control Trials: The Case of a Training Program to Encourage Social Entrepreneurship

2020 ◽  
Author(s):  
Thomas Åstebro ◽  
Florian Hoos
Keyword(s):  
Training Program ◽  
Social Entrepreneurship ◽  
Randomized Control Trials ◽  
Impact Measurement ◽  
Randomized Control

EXPRESS: Comparison of Mothership versus Drip-and-Ship Models in treating patients with Acute Ischemic Stroke: A systematic review and meta-analysis

2021 ◽  
pp. 174749302110132
Author(s):  
Ahmed Mohamed ◽  
Nida Fatima ◽  
Ashfaq Shuaib ◽  
Maher Saqqur
Keyword(s):  
Functional Outcome ◽  
Meta Analysis ◽  
Critical Time ◽  
Functional Independence ◽  
Models Of Care ◽  
Randomized Control Trials ◽  
Symptomatic Intracerebral Hemorrhage ◽  
Significant Difference ◽  
The Mean ◽  
Randomized Control

Introduction There is controversy if direct to comprehensive center “mothership” (MS) or stopping at primary center for thrombolysis before transfer to comprehensive center “drip-and- ship” (DS) are best models of treatment of acute stroke. In this study, we compare MS and DS models to evaluate the best option of functional outcome. Methods Studies between 1990 and 2020 were extracted from online electronic databases. We compared the clinical outcomes, critical time measurements, functional independence and mortality were then compared. Results A total of 7,824 patients’ data were retrieved from 13 publications (3 randomized control trials and 10 retrospective ones). 4,639 (59.3%) patients were treated under MS model and 3,185 (40.7%) followed the DS model with mean age of 70.01±3.58 vs. 69.03±3.36; p< 0 .001, respectively. The National Institute Health Stroke Scale was 15.57±3.83 for the MS and 15.72±2.99 for the DS model (p=<0.001). The mean symptoms onset-to-puncture time was significantly shorter in the MS group compared to the DS (159.69 min vs. 223.89 min; p=<0.001, respectively). Moreover, the collected data indicated no significant difference between symptom’s onset to intravenous (IV) thrombolysis time and stroke onset-to-successful recanalization time (p=0.205 and p=<0.001, respectively). Patients had significantly worse functional outcome [modified rankin score (mRS)] (3-6) at 90-days in the DS model [Odds Ratio (OR): 1.47, 95% Confidence Interval (CI): 1.13-1.92, p<0.004] and 1.49-folds higher likelihood of symptomatic intracerebral hemorrhage (OR: 1.49, 95%CI: 1.22-1.81, p<0.0001) compared to MS. However, there were no statistically significant difference in terms of mortality (OR: 1.16, 95%CI: 0.87-1.55, p=0.32) and successful recanalization (OR: 1.12, 95%CI: 0.76-1.65, p=0.56) between the two models of care. Conclusion Patients in the MS model have significantly improved functional independence and recovery. Further studies are needed as the data from prospectively randomized studies is not of sufficient quality to make definite recommendations.

Site Selection Bias in Program Evaluation *

2015 ◽  
Vol 130 (3) ◽  
pp. 1117-1165 ◽  
Author(s):  
Hunt Allcott
Keyword(s):  
United States ◽  
Selection Bias ◽  
Site Selection ◽  
The United States ◽  
Randomized Control Trials ◽  
Program Evaluations ◽  
Out Of Sample ◽  
Target Treatment ◽  
Randomized Control ◽  
Selection For

Abstract “Site selection bias” can occur when the probability that a program is adopted or evaluated is correlated with its impacts. I test for site selection bias in the context of the Opower energy conservation programs, using 111 randomized control trials involving 8.6 million households across the United States. Predictions based on rich microdata from the first 10 replications substantially overstate efficacy in the next 101 sites. Several mechanisms caused this positive selection. For example, utilities in more environmentalist areas are more likely to adopt the program, and their customers are more responsive to the treatment. Also, because utilities initially target treatment at higher-usage consumer subpopulations, efficacy drops as the program is later expanded. The results illustrate how program evaluations can still give systematically biased out-of-sample predictions, even after many replications.

Sign in / Sign up

Export Citation Format

Share Document