ABSTRACT
Un ensayo clínico aleatorizado por conglomerados se da cuando se aleatorizan grupos (clusters) de individuos a las distintas ramas. Puede ser la única o mejor opción de diseño ante determinadas circunstancias: si hay un claro agrupamiento (biológico o funcional) en donde algunos individuos de análisis son más parecidos entre sí que otros; si las intervenciones a evaluar se realizan a nivel del conglomerado; cuando hay riesgo de contaminación; o por practicidad, costos o conveniencia. Entre los problemas más importantes que conllevan se encuentran posibles sesgos (especialmente cuando el reclutamiento de los individuos se realiza luego de la aleatorización, o no existe ceguera), así como mayor complejidad en el diseño y análisis. Asimismo, si no se tienen en cuenta la agrupación de individuos por conglomerados para el cálculo del tamaño muestral o del análisis de los datos, se podrían obtener resultados incorrectos. Estos estudios deben explicitar, además de lo habitualmente reportado: por qué se decidió realizar un diseño por conglomerados; si los objetivos, intervenciones y puntos finales a evaluar apuntan a nivel del conglomerado, individual, o ambos; describir los criterios de inclusión a nivel del conglomerado e individual; mostrar cómo se hicieron el cálculo del tamaño muestral y los análisis considerando los conglomerados; aclarar si los pacientes, profesionales actuantes e investigadores estaban ciegos a las ramas de investigación; y discutir la generalizabilidad de los resultados, entre otros. Si bien tienen mayor complejidad, estos estudios son cada vez más frecuentes. Es un diseño muy útil si está bien desarrollado y es importante conocer sus particularidades. (AU)
We perform a cluster randomized controlled trial when we randomize groups (or clusters) of individuals (whether humans, cells, or clinics) to different study arms, and not simply individuals. It can be the only or best study design option in certain circum-stances: if there is a clear grouping, when some subjects of analysis are more similar among them than the rest; if interventions to be evaluated are made at cluster level; when there is risk of "contamination" or cross-over; or because of practicality, costs or convenience according to researchers judgment. Cluster trials are associated with important issues: risk of bias (especially when individuals recruitment is made after randomization, or if there was no blinding); and the need of more complex design and analysis. If we do not take clusters into account in the sample size estimation and data analysis, we could get misleading results.When reporting these studies, researchers should make explicit (in addition to standard reporting requirements): the rationale for a cluster design; if the objectives, interventions and endpoints are for clusters, individuals or both; the inclusion criteria for clusters and individuals; how they did sample size estimations and data analysis considering cluster design; if patients, health care profes-sionals and researchers were blind; and if results can be generalized. Even though cluster randomized controlled trials are more complex, these studies are increasingly common. It is a very useful design, if correctly done. And it is important to understand its main characteristics. (AU)
Subject(s)
Humans , Randomized Controlled Trials as Topic/methods , Cluster Sampling , Cluster Analysis , Randomized Controlled Trials as Topic/classification , Randomized Controlled Trials as Topic/ethics , Selection Bias , Epidemiologic Study CharacteristicsABSTRACT
OBJECTIVE:We refer to the effectiveness (known as pragmatic or real world) and efficacy (known as explanatory or desired or ideal world) of interventions. However, these terms seem to be randomly chosen by investigators who design clinical trials and do not always reflect the true purpose of the study. A pragmatic-explanatory continuum indicator summary tool was thus developed with the aim of identifying the characteristics of clinical trials that distinguish between effectiveness and efficacy issues. We verified whether clinical trials used the criteria proposed by the indicator summary tool, and we categorized these clinical trials according to a new classification.METHOD:A systematic survey of randomized clinical trials was performed. We added a score ranging from 0 (more efficacious) to 10 (more effective) to each domain of the indicator summary tool and proposed the following classifications: high efficacy (<25), moderate efficacy (25-50), moderate effectiveness (51-75), and high effectiveness (<75).RESULTS:A total of 844 randomized trials were analyzed. No analyzed trials used the criteria proposed by the indicator summary tool. Approximately 44% of the trials were classified as having moderate effectiveness, and 43.82% were classified as having moderate efficacy.CONCLUSIONS:Most clinical trials used the term “efficacy” to illustrate the application of results in clinical practice, but the majority of those were classified as having moderate effectiveness according to our proposed score. The classification based on the 0-100 score is still highly subjective and can be easily misunderstood in all domains based on each investigator’s own experiences and knowledge.
Subject(s)
Humans , Randomized Controlled Trials as Topic/statistics & numerical data , Treatment Outcome , Bibliometrics , Clinical Trials as Topic/classification , Research Design , Randomized Controlled Trials as Topic/classificationABSTRACT
OBJECTIVE: To evaluate the validity of the Qualis database in identifying the levels of scientific evidence and the quality of randomized controlled trials indexed in the Lilacs database. METHODS: We selected 40 open-access journals and performed a page-by-page hand search, to identify published articles according to the type of study during a period of six years. Classification of studies was performed by independent reviewers assessed for their reliability. Randomized controlled trials were identified for separate evaluation of risk of bias using four dimensions: generation of allocation sequence, allocation concealment, blinding, and incomplete outcome data. The Qualis classification was considered to be the outcome variable. The statistical tests used included Kappa, Spearman's correlation, Kendall-tau and ordinal regressions. RESULTS: Studies with low levels of scientific evidence received similar Qualis classifications when compared to studies with high levels of evidence. In addition, randomized controlled trials with a high risk of bias for the generation of allocation sequences and allocation concealment were more likely to be published in journals with higher Qualis levels. DISCUSSION: The hierarchy level of the scientific evidence as classified by type of research design, as well as by the validity of studies according to the bias control level, was not correlated or associated with Qualis stratification. CONCLUSION: Qualis classifications for journals are not an approximate or indirect predictor of the validity of randomized controlled trials published in these journals and are therefore not a legitimate or appropriate indicator of the validity of randomized controlled trials.