Design and Analysis of Group-Randomized Trials: A Review of Recent Methodological Developments

Murray, David M.; Varnell, Sherri P.; Blitstein, Jonathan

doi:10.2105/ajph.94.3.423

Cited by 562 publications

(497 citation statements)

References 123 publications

Supporting

Mentioning

491

Contrasting

Unclassified

Order By: Relevance

“…The gold standard for inference in community interventions is a community-randomized trial because the design eliminates confounding bias (2). Bias from other sources can result from frequent measurement (3) or lack of blinding treatment (blinding is rarely possible for community interventions) (4,5).…”

mentioning

confidence: 99%

Causal inference methods to study nonrandomized, preexisting development interventions

Arnold

Khush

Padmavathi

et al. 2010

Proc. Natl. Acad. Sci. U.S.A.

View full text Add to dashboard Cite

Empirical measurement of interventions to address significant global health and development problems is necessary to ensure that resources are applied appropriately. Such intervention programs are often deployed at the group or community level. The gold standard design to measure the effectiveness of communitylevel interventions is the community-randomized trial, but the conditions of these trials often make it difficult to assess their external validity and sustainability. The sheer number of community interventions, relative to randomized studies, speaks to a need for rigorous observational methods to measure their impact. In this article, we use the potential outcomes model for causal inference to motivate a matched cohort design to study the impact and sustainability of nonrandomized, preexisting interventions. We illustrate the method using a sanitation mobilization, water supply, and hygiene intervention in rural India. In a matched sample of 25 villages, we enrolled 1,284 children <5 y old and measured outcomes over 12 mo. Although we found a 33 percentage point difference in new toilet construction [95% confidence interval (CI) = 28%, 39%], we found no impacts on height-for-age Z scores (adjusted difference = 0.01, 95% CI = −0.15, 0.19) or diarrhea (adjusted longitudinal prevalence difference = 0.003, 95% CI = −0.001, 0.008) among children <5 y old. This study demonstrates that matched cohort designs can estimate impacts from nonrandomized, preexisting interventions that are used widely in development efforts. Interpreting the impacts as causal, however, requires stronger assumptions than prospective, randomized studies.impact evaluation | study design | propensity score matching | communityled total sanitation | open defecation

show abstract

mentioning

confidence: 99%

Causal inference methods to study nonrandomized, preexisting development interventions

Arnold

Khush

Padmavathi

et al. 2010

Proc. Natl. Acad. Sci. U.S.A.

View full text Add to dashboard Cite

show abstract

“…This study has as framework a generalized linear model where the dependent variable is a numeric variable; which it is supported by current research, indicating the utility of such approaches for studies in public health with this kind of features [79][80][81][82][83][84]. There are some critics that mention possible risks and limitations with this approach.…”

Section: Discussionmentioning

confidence: 55%

Motorcyclists’ Mortality Pattern in Colombia from 2000 to 2013: A Longitudinal Study

Martin¹,

Tovar²,

Cidronio³

et al. 2017

Arch Med

View full text Add to dashboard Cite

show abstract

“…This model is being increasingly appreciated and used by social scientists who are keen to improve the rigour of the evidence that randomized experiments can produce, both in the field (Duflo, 2008;Kremer and Glennester, 2012) and in the lab (Murray et al, 2004;Falk and Heckman, 2009). However, in many areas of social sciences the value of experiments is still debated for at least two major reasons.…”

Section: The Value and Limits Of The Medical Model Of Nutrition Intermentioning

confidence: 99%

Health, Wellbeing and Social Sciences

Fattore

Agostoni

2015

Critical Reviews in Food Science and Nutrition

View full text Add to dashboard Cite

For social interventions aimed at improving nutrition behavior evidence from randomized trials is essential but cannot be the only approach of research activities. Interventions on dietary habits require considerations on food security, economic and environmental sustainability, and a broad meaning of wellbeing which includes, but also goes beyond, health effects. The model of research in nutrition requires a new consideration of observational studies, mainly through different analytical models. Nutrition and food studies need research programs where medical (nutrition and health), psychology (how we behave), economics (how resources are used and their impact on wellbeing) and sociology (how social determinant shape behavior) collaborate.

show abstract

Design and Analysis of Group-Randomized Trials: A Review of Recent Methodological Developments

Cited by 562 publications

References 123 publications

Causal inference methods to study nonrandomized, preexisting development interventions

Causal inference methods to study nonrandomized, preexisting development interventions

Motorcyclists’ Mortality Pattern in Colombia from 2000 to 2013: A Longitudinal Study

Health, Wellbeing and Social Sciences

Contact Info

Product

Resources

About