Some quotes about statistics or from statisticians

I often see people cite interesting quotes in their presentation. Citing a good quote can entice the audience. I recently saw a presentation with the following quote about the statistics in dealing with uncertainties:

One way of defining statistics is…
The science of quantifying uncertainty, dealing with uncertainty, and making decisions in the face of uncertainty…
…and drug development is a series of decisions under huge uncertainty.

There is a website that lists the famous quotes from Statisticians or about Statisticians

Here are some quotes with top ranking:  

All models are wrong, but some are useful. (George E. P. Box)

An approximate answer to the right problem is worth a good deal more than an exact answer to an approximate problem. (John Tukey)

To call in the statistician after the experiment is done may be no more than asking him to perform a post-mortem examination: he may be able to say what the experiment died of. (Ronald Fisher (1938))

Statistics are like bikinis. What they reveal is suggestive, but what they conceal is vital. (Aaron Levenstein)

Statisticians, like artists, have the bad habit of falling in love with their models. (George Box)

I think it is much more interesting to live with uncertainty than to live with answers that might be wrong. (Richard Feynman)

If you torture the data enough, nature will always confess. (Ronald Coase)

Absence of evidence is not evidence of absence. (Carl Sagan)

It's easy to lie with statistics; it is easier to lie without them. (Frederick Mosteller)

Then there is a following cartoon::

New books about 'missing data in clinical trials'

Missing data is one of the critical issues in statistics and in clinical trials. Three new books about the missing data were written by the biostatisticians working directly in pharmaceutical and drug development fields. These three books are worth recommending.

• Michael O'Kelley and Baohdana Ratitch (2014) Clinical Trials with Missing Data: A guide for Practitioners - also see O'kelley's recent talk "Missing data in clinical trials: developments in practice"

• Craig H. Mallinckrodt (2013) Preventing and Treating Missing Data in Longitudinal Clinical Trials

• James R. Carpenter, Michael G. Kenward (2013) Multiple Imputation and Its Application (Statistics in Practice) 

For free books regarding the missing data, the following resources are available: 

• Panel on Handling Missing Data in Clinical Trials Committee on National Statistics, Division of Behavioral and Social Sciences and Education, NATIONAL RESEARCH COUNCIL (2010) The Prevention and Treatment of Missing Data in Clinical Trials

• James R. Carpenter & Michael G. Kenward (2007) Missing data in randomised controlled trials — a practical guide

Does FDA or EMA have a preferred method for dealing with missing data for when companies submit new drugs for approval?

• The EMA has a guideline document on the handling of missing data in trials titled "Guideline on Missing Data in Confirmatory Clinical Trials" 
• The FDA does not have a single document which contains its recommendations for handling missing data in clinical trials. It appears that the FDA does not plan to issue a new guidance on missing data any time soon. However, the FDA often cited NRC’s “The Prevention and Treatment of Missing Data in Clinical Trials” for directions in dealing with the missing data. 

The general trends in handling the missing data:

• Shifting from handling the missing data to preventing the missing data
• Single imputation methods are being criticized
• None of the missing data handling approach is perfect (for example, MMRM method is criticized in recent FDA advisory committee meeting)
• New missing data handling approaches are proposed and used (for example, the jump to reference (or J2R) imputation approach used by FDA statistician in Pulmonary-Allergy Drugs Advisory Committee Meeting September 10, 2013)
• When missing data becomes an issue, sensitivity analyses with different imputation approach should be on must-to-do list especially for phase III studies – it will be a problem if the contradictory results are found with different imputation methods. See Olanzapine NDA review memos