|
|
|
|
|
|
|
|
Category: Early stopping in clinical trials. Clinical trials are sometimes stopped because of early evidence of efficacy, early evidence of harm, or early evidence of futility. In general, the rules for stopping a study need to be specified in the research protocol before any data is collected. These pages discuss some of the issues associated with early stopping of clinical trials. Articles are arranged by date with the most recent entries at the top. You can find the theme and closely related categories and other resources at the bottom of this page.
Stats: What does one-third of the way through a study mean? (April 6, 2008). Someone asked me a very good question regarding interim analysis. If the call for an interim analysis is specified as occuring one-third (and two-third) of the way through a study, what does that mean. In a study with 60 patients lasting a full year, does that mean until 20 (40) patients have arrived, or does it mean until 4 (8) months? Also, if you are counting discrete events, such as adverse events, does it mean one-third of the expected number of adverse events?
Stats: When are two events worrisome? (December 4, 2006). I was attending a workshop on developing a risk management plan for a new drug. The presenter, Nawab Qizilbash from Oxon Clinical Epidemiology, gave an interesting example. Suppose you are running a clinical trial with 50 total patients. In the treatment group, you notice two adverse events and in the placebo group you notice zero adverse events. Should you stop the trial?
Stats: Oops! I found statistical significance in my pilot study (September 18, 2006). Dear Professor Mean, I ran a pilot study to help me estimate the sample size for a larger study. There were 15 patients in each arm of the pilot, and when I analyzed the pilot data for efficacy in the primary outcome, I was shocked to find out that the results were already statistically significant. Do I still need to run the full study?
Stats: Stopping rules for a Data Safety Monitoring Plan (July 5, 2006). If a research study requires a DSMP (Data Safety Monitoring Plan), that plan should outline conditions that would cause a study to end early. It is difficult to specify what those conditions would be, but it is important to at least think about and comment on each of the major areas listed below.
Stats: Group Sequential Monitoring of Clinical Trials in R (December 13, 2005). It is very expensive to purchase software that performs group sequential monitoring of clinical trials (sometimes called interim analysis). Group sequential monitoring is looking at a trial at selected time points during the study to see if you should stop the study early. There are a couple of functions in R that will do simple calculations, and the price, of course, is free.
Stats: When can I stop my CQI study? (June 6, 2005). I was asked today about a CQI project where children were being tested for a certain condition, but they never tested positive. This was going on for over a year and there were already 188 cultures and none were positive. Is it safe to hang up your spurs and call it a day?
Stats: Do I have enough data after 24 months of time? (April 5, 2005). Someone asked me about a correlation coefficient that he computed on a data set that represented 24 months of data collection. A particular correlation of interest (a correlation between staff turnover and resident falls) was not significantly different from zero, but this person wanted to know how much more data to collect before safely concluding that no relation has been or likely will be established. First compute a confidence interval for the correlation coefficient. If that interval is so narrow that you can rule out the possibility of a clinically important shift, then your sample size is large enough.
Stats: Controversy over stopping a study early (November 24, 2004). A while back, the IRB asked me to look into a randomized study where the interim report indicated a huge disparity in the two treatment arms. One arm of the study had almost all good outcomes and the other arm had almost all bad outcomes or at best no improvement. The sample size, though, was only 20 patients, and the protocol had no formal rule for stopping the study early. Even without such a rule, a careful analysis of the data revealed that there was little justification for continuing randomization when one arm of the study was clearly inferior.
Stats: Early stopping in an animal experiment (July 1, 2004). I was helping someone with power calculations for an animal experiment. There were several ways to estimate power and they all seemed to point to the use of either 8 animals per group or 12 animals per group. I explained that if she was more risk averse, she should use 12 animals and if she were more work averse, she should use 8 animals. Her answer surprised me a bit. She said that she liked the idea of using 12 animals per group. That way, if the results achieved significance after only 8 animals, she could go back to the Animal Care and Use Committee and tell that she saved a few animals.
Stats: Stopping a study early (October 29, 2002). Dear Professor Mean, I tried really hard to recruit the number of subjects that I promised to in my power calculations, but I just can't do it. I'm thinking about stopping the study early, but I'm worried that it might screw up all my statistics. -- Exhausted Evelyn
Stats: Protocol changes (December 22, 2000). Dear Professor Mean, After I collected my data, I noticed a problem that I had not anticipated. I want to make some protocol changes and analyze my data differently. Can I do this?
Stats: Interim analysis (September 13, 1999). Dear Professor Mean, I'm going on a job interview and I know one of the questions they will ask me is about interim analysis. What should I tell them? -- Harried Howard
Theme and closely related categories:
Examples
[Return to full topic list] [Read current weblog entries]
This webpage was written by Steve Simon on 2007-06-18 and was last modified on 2008-07-14. Send feedback to ssimon at cmh dot edu or click on the email link at the top of the page.