#9 PH150A Feb8 Randomized Control Trials

Often appropriate to have several outcomes often

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: en appropriate to have several outcomes Often appropriate to have several outcomes Loss followLoss to follow-up (attrition) can cause misleading result Optimize Optimize participation (same techniques as cohort study) Assess Assess whether drop-outs differed in outcome from dropthose who completed study Can also analyze data “assuming the worst”-that all th th drop drop outs developed the outcome. 20 Anatomy of an Experimental Study Anatomy of an Experimental Study Intervention Intervention Group Random Assignment best Rate of Outcom e No Intervention Control Group Rate of Outcom e Start with healthy people for preventive trial Control group may receive Sick people for therapeutic trial nothing, placebo, usual treatment etc treatment etc Present Future 21 Validity Validity Results of study are valid when they Results of a study are valid when they reflect reflect the truth. Internal validity refers to the design methods Internal validity refers to the design, methods, analysis and results of an individual study. Is the comparison between study groups valid (free (free of systematic error/bias)? External External validity refers to whether internally valid findings are also applicable lid fi (generalizable) to people other than those generalizable) included in an individual study. 22 Internal Internal and External External Validity in Validity in a Randomized Randomized Trial Trial Gordis, page 153 (4th 153 (4 ed) 23 Efficacy and Effectiveness Efficacy and Effectiveness Efficacy Efficacy: extent to which an intervention produces beneficial results under ideal conditions. id Analyze Analyze only people that adhered to the study protocol (assigned treatment) Effectiveness Effectiveness: extent to which an intervention produces beneficial results under usual (real world) conditions. conditions. Analyze people as they were randomized, whether or not they adhered to study protocols See See Gordis text (pg 294 and 295) 24 Intention Intention-to-treat Analysis Analysis Efficacy Efficacy study Does the intervention have an effect in those Does the intervention have an effect in those who who actually receive it? it? Thi...
View Full Document

This note was uploaded on 02/21/2010 for the course PH 150A taught by Professor Adams during the Spring '08 term at Berkeley.

Ask a homework question - tutors are online