Research design considerations for confirmatory chronic pain clinical trials: IMMPACT recommendations.

There has been an increase in the number of chronic pain clinical trials in which the treatments being evaluated did not differ significantly from placebo in the primary efficacy analyses despite previous research suggesting that efficacy could be expected. These findings could reflect a true lack of efficacy or methodological and other aspects of these trials that compromise the demonstration of efficacy. There is substantial variability among chronic pain clinical trials with respect to important research design considerations, and identifying and addressing any methodological weaknesses would enhance the likelihood of demonstrating the analgesic effects of new interventions. An IMMPACT consensus meeting was therefore convened to identify the critical research design considerations for confirmatory chronic pain trials and to make recommendations for their conduct. We present recommendations for the major components of confirmatory chronic pain clinical trials, including participant selection, trial phases and duration, treatment groups and dosing regimens, and types of trials. Increased attention to and research on the methodological aspects of confirmatory chronic pain clinical trials has the potential to enhance their assay sensitivity and ultimately provide more meaningful evaluations of treatments for chronic pain.

Attrition in randomized controlled clinical trials: methodological issues in psychopharmacology.

Attrition is a ubiquitous problem in randomized controlled clinical trials (RCT) of psychotropic agents that can cause biased estimates of the treatment effect, reduce statistical power, and restrict the generalizability of results. The extent of the problem of attrition in central nervous system (CNS) trials is considered here and its consequences are examined. The taxonomy of missingness mechanisms is then briefly reviewed in order to introduce issues underlying the choice of data analytic strategies appropriate for RCTs with various forms of incomplete data. The convention of using last observation carried forward to accommodate attrition is discouraged because its assumptions are typically inappropriate for CNS RCTs, whereas multiple imputation strategies are more appropriate. Mixed-effects models often provide a useful data analytic strategy for attrition as do the pattern-mixture and propensity adjustments. Finally, investigators are encouraged to consider asking participants, at each assessment session, the likelihood of attendance at the subsequent assessment session. This information can be used to eliminate some of the very obstacles that lead to attrition, and can be incorporated in data analyses to reduce bias, but it will not eliminate all attrition bias.

A Comparison of Filter Types in the Collection and Gravimetric Determination of Airborne Particulate Matter Less than 2.5 Microns (PM25).

Five identical, collocated, low-volume samplers were operated to collect airborne particulate matter less than 2.5 microns (PM25). Five commercially available filter types were installed in the samplers to compare the gravimetric determination of PM2.5 concentrations in the atmosphere. The filters were rotated through the five samplers for two study periods-one in summer and one in winter. The study was performed in Sheridan, WY, in close proximity to a gravimetric laboratory to minimize the introduction of errors associated with sample handling. Rigorous quality assurance procedures were employed throughout the study.Four of the five filter types provided comparable gravimetric determinations of airborne PM2.5.