Issues in data monitoring and interim analysis of trials. AM Grant, DG Altman, AB
Babiker, MK Campbell, FJ Clemens, JH Darbyshire, DR Elbourne, SK McLeer, MKB Parmar, SJ
Pocock, DJ Spiegelhalter, MR Sydes, AF Walker, SA Wallace. Health Technolgoy Assessment 2005:
9(7);
[Medline] [Abstract]
[PDF]
Offering results to research participants. E. M. Dinnett, M. M. Mungall, C. Gordon,
E. S. Ronald, A. Gaw. Bmj 2006: 332(7540); 549-50.
[Medline] [Excerpt] The distinction between giving a general summary of trial results
to study participants and providing them with their own trial results is not made clear in
the paper by Dixon-Woods et al or the accompanying editorial. Investigators may provide
patients with a summary of the trial results, but even if they do not they should be aware
that study participants may access these results elsewhere. The provision of personalised
trial results, and in particular treatment allocation (unblinding), is quite a different
matter and much debated. This is the more likely of the two to result in emotional
consequences for the participants.
One-time general consent for research on biological samples. D. Wendler. Bmj 2006:
332(7540); 544-7.
[Medline] [Excerpt] It is now recognised that people should give informed consent for
use of their biological samples in research. The literature on individuals' views supports
one-time general consent as the best approach for this purpose.
Wonder Drug Inspires Deep,
Unwavering Love Of Pharmaceutical Companies. The Onion, Published in Issue
42-10, March 6, 2006. Accessed on 2006-03-09. [Excerpt] The Food and Drug Administration
today approved the sale of the drug PharmAmorin, a prescription tablet developed by Pfizer to
treat chronic distrust of large prescription-drug manufacturers. Pfizer executives
characterized the FDA's approval as a "godsend" for sufferers of independent-thinking-related
mental-health disorders. www.theonion.com/content/node/46032
Flame Warriors.
Mike Reed. Accessed on 2006-03-08. [Excerpt] Some years ago, a minor pat ignited a searing
falme war that threatened to consume a one-placid discussion forum. While the forum burned, I
amused myself by caricaturing the chief antagonists. Confounded at seeing themselves thus
revealed, the combatants fled the field in disarray. Over time the roster of online
belligerents expanded and eventually congealed into the netizen's guide to Flame Warriors. My
own bad internet behavior would certainly have provided sufficient material to populate an
extensive rogue's gallery, but suggestions and comments from astute observers continue to
enrich the Flame Warriors collection. Please report immediately any sightings of new Warrior
variants. redwing.hutman.net/~mreed/index.htm
A Regression Model for Dependent Gap Times. Robert L. Strawderman. The
International Journal of Biostatistics 2006: 2(1); 1-34.
[Abstract] A natural choice of time
scale for analyzing recurrent event data is the "gap" (or soujourn) time between successive
events. In many situations it is reasonable to assume correlation exists between the
successive events experienced by a given subject. This paper looks at the problem of
extending the accelerated failure time (AFT) model to the case of dependent recurrent event
data via intensity modeling. Specifically, the accelerated gap times model of Strawderman
(2005), a semiparametric intensity model for independent gap time data, is extended to the
case of multiplicative gamma frailty. As argued in Aalen & Husebye (1991), incorporating
frailty captures the heterogeneity between subjects and the "hazard" portion of the intensity
model captures gap time variation within a subject. Estimators are motivated using
semiparametric efficiency theory and lead to useful generalizations of the rank statistics
considered in Strawderman (2005). Several interesting distinctions arise in comparison to the
Cox-Andersen-Gill frailty model (e.g., Nielsen et al, 1992; Klein, 1992). The proposed
methodology is illustrated by simulation and data analysis.
ZumaStat. James Jaccard.
Accessed on 2006-03-03. [Excerpt] ZumaStat offers stand alone statistical programs that
can be integrated easily with the menu bars of SPSS and Excel. The programs work with summary
statistics (such as means, standard deviations, percentages, and correlations) to perform a
wide range of statistical tests and power analyses (e.g., test of difference between two
percentages, test of difference between two correlations, single degree of freedom
interaction contrasts, t tests for means). ZumaStat also has a special SPSS utility (called Z
Plus) that makes using SPSS much easier and more flexible. www.zumastat.com/Home.htm
Understanding the design matrix
in linear models for microarray experiments. Natalie Thorne, University of
Cambridge, The Hutchison/MRC Research Center. Accessed on 2006-03-03. www.damtp.cam.ac.uk/user/npt22/
These two articles have not been published yet, but I will track down the information when
they appear:
-
Djulbegovic, Clarke: Scientific and ethical issues in equivalence trials. JAMA March 7,
2001;285:1206-1208
-
Morse, Califf, Sugarman: Monitoring and ensuring safety during clinical research. JAMA March
7, 2001;285:1201-1205
Monitoring and ensuring safety during clinical research. M. A. Morse, R. M. Califf,
J. Sugarman. Jama 2001: 285(9); 1201-5.
[Medline] [Abstract]
[Full text]
[PDF] Increased numbers
of clinical trials, many of which are large, multicenter, and sometimes international, and
the marked shift of funding for clinical trials to industry have made apparent the inadequacy
of mechanisms for protecting human subjects that were developed when clinical research was
generally carried out on a small scale at single institutions. To address concerns regarding
the protection of human subjects, a group of professionals with expertise in various aspects
of clinical trials was assembled in May 2000. Participants described and evaluated the
mechanisms by which clinical trials are monitored, focusing on adverse event reporting and
the processes by which various parties with oversight responsibilities interact in the course
of these trials. In this article, we describe the manner in which adverse event reporting
might function to enhance safety and the role of data monitoring committees in using
aggregate data from these reports, outline the problems that now exist for institutional
review boards as they are faced with multiple adverse event reports from clinical trials
while conducting continuing review, and offer recommendations for improving the current
approach.
Analysis of
Covariance Primer [PDF]. Dennis Roberts, Penn State University. Accessed on
2006-03-02. [Excerpt] In the normal experimental design, Ss are randomly assigned to
different treatment levels. and then we might perform a simple ONE factor ANOVA to test the
null hypothesis of no treatment effects in the overall populations. Consider the simplest of
these 'experiments', a two group study where 20S have been randomly assigned (n=10 to each
group) to two conditions: Experimental and Control. In the data table below, YE means the
dependent variable data (% correct scores on a science test) for the Experimental group and
YC is the comparable data for the Control group. With the DESC output, you will note that the
mean E is 68.1 and the mean C is 61. Then question here of course is whether we would, after
our statistical test, reject the null hypothesis. www.personal.psu.edu/users/d/m/dmr/papers/ANCOVA1.PDF
Guidance on
Research Involving Coded Private Information or Biological Specimens [PDF].
Office for Human Research Protections, Department of Health and Human Services, Published
August 10, 2004. Accessed on 2006-03-02. (Ethics, Privacy) [Excerpt] Scope: This
document applies to research involving coded private information or human biological
specimens (hereafter referred to as 'specimens') that is conducted or supported by HHS. This
document does the following: (1) Provides guidance as to when research involving coded
private information or specimens is or is not research involving human subjects, as defined
under HHS regulations for the protection of human research subjects (45 CFR part 46). (2)
Reaffirms OHRP policy (see OHRP guidance on repository activities and research on human
embryonic stem cells) that, under certain limited conditions, research involving only coded
private information or specimens is not human subjects research. (3) Clarifies the
distinction between (a) research involving coded private information or specimens that does
not involve human subjects and (b) human subjects research that is exempt from the
requirements of the HHS regulations. (4) References pertinent requirements of the HIPAA
Privacy Rule that may be applicable to research involving coded private information or
specimens. www.hhs.gov/ohrp/humansubjects/guidance/cdebiol.pdf
HRPP Blog. Jeffrey Cohen.
Accessed on 2006-03-02. [Excerpt] The purpose of this blog is to share my experience with
human research protections programs. hrpp.blogspot.com
07/08/2008.
