#9 PH150A Feb8 Randomized Control Trials

9 time time tunnel for study designs survey cross

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: l for Study Designs Survey: Cross Sectional Study Prospective Cohort Study Retrospective Cohort Study [ Case Control Study St Nested Case Control Study Case Control Study Intervention Study PAST PRESENT FUTURE 10 1. Select Study Group 1. Select Study Group Use clear, practical, standardized criteria. Use Ideally, individuals in group must: Not have outcome being studied Be Be at risk of developing the outcome Be Be representative of target population Be Be likely to benefit from intervention Be willing and able to be randomized and comply with intervention Be able to be followed to assess for development Be able to be followed to assess for development of of outcome 11 Special Special issues of informed consent consent Is randomization ethical? Is use of placebo Is randomization ethical? Is use of a placebo ethical? ethical? When is it ethical to withhold treatment? Is it ethical to not randomize? Is it ethical to not randomize? Can Can informed consent truly be obtained? When When should the trial be stopped early either due to harmful or beneficial findings? (A (A separate ethical issue: role of funder in conducting, interpreting and reporting results) 12 2. 2. Assign to Treatment and and Control Groups Comparison groups should come from the same th underlying population as the treatment group and ideally be studied concurrently ideally be studied concurrently. Types of comparison groups Randomized Randomized Simultaneous Simultaneous nonrandomized controls Historical Historical controls (“before-after” study) (“beforeSubject as own control (“cross-over” design) Sub “cross13 http://individual.utoronto.ca/coppin/exp erimental_group_larger_edge800w.jpg 14 Randomization Randomization Decreases probability of confounding bias Decreases Usually Usually equalizes differences between groups Randomize Randomize AFTER participants are screened Confirm Confirm that randomized groups are balanced with regard to key characteristics http://individual.utoronto.ca/coppin/...
View Full Document

Ask a homework question - tutors are online