Journal Article

The use of Latin squares and related block designs in implementation research

Eccles, Martin P., Campbell, Marion K., Ramsay, Craig R., Russell, Ian T., Steen, Ian Nicholas, Grimshaw, Jeremy M.

2014

Overview

In practice, the implementation studies that have used this design have involved multiple centers in each block; by modeling differences between centers as random effects rather than fixed effects, it is possible to obtain interval estimates of treatment effects (see Appendix on journal's website at www.jclinepi.com). Analyzing the two conditions separately estimates effectiveness only through the first of these comparisons; analyzing conditions simultaneously uses more of the available information but requires a more restrictive set of assumptions about the observed data [4] (see Appendix on journal's website at www.jclinepi.com).\\n Increased applicability of block designs in implementation research BIB and Latin square designs have been used to evaluate alternative implementation strategies while controlling for one or more nuisance variables [6-14].

• Publisher: Elsevier Inc,Elsevier BV,Elsevier,Elsevier Limited
• Subject: Analysis. Health state / Biological and medical sciences / Biomedical Research / Biomedical Research - methods / Cardiovascular disease / Epidemiology / Estimates / General aspects / Generalized linear models / Humans / Internal Medicine / Medical sciences / Public health. Hygiene / Public health. Hygiene-occupational medicine / R / R Medicine / Research Design / Studies / Variables / Websites