Article type
Year
Abstract
Introduction: The uncertainty principle represent a fundamental scientific and ethical principle for conducting randomized controlled trials (RCTs) (BMJ 1998;317:1170). It states that the patient should be enrolled in RCTs only if there is substantial uncertainty which of the trial treatments would benefit a patient. Very little work has been done to demonstrate whether this theoretical principle is observed in actual practice. If the uncertainty principle holds true, we would expect to find no significant difference in the number of successful results on experimental treatments vs. standard ones.
Objectives:
Methods: We focused on one disease category (multiple myeloma). We first sought all RCTs for this disease from 1996-1998, identifying 132 RCTs. To evaluate which treatment is better in a given RCT, we followed the intention of the original investigators and recorded them in the manner reported. The results of the each study were rated on a six point scale (1-standard treatment highly preferred, 6-experimental treatment highly preferred) with a value of 3.5 representing no difference between the two types of therapy (the point of equipoise). The features of RCTs which might influence reported results such as the type of treatment and the source of funding were studied. The rating of studies was performed by a single rater (ML) who was blinded to the source of funding. About 10-20% of all identified trials were also independently reviewed by the first author. Statistical inference on subgroups was performed utilizing the Mann-Whitney rank sum test and chi-square test for comparison of rankings and proportions, respectively
Results: Overall, the uncertainty principle was found to hold true in the conduct of RCTs in multiple myeloma. The median evaluation score for all 132 RCTs was 3.5 (63% of RCTs), ranging from a score of 2 (3%) to a score 6 (0.75%; only 1 RCT had score of 6). When the data were analyzed according to the type of treatment, the following scores (median [range]) were observed: induction 3.5 [2 to 5], maintenance 3.5 [3.5 to 5], supportive care 4.5 [3.5 to 6], salvage 3.5 [3.5 to 3.5], consolidation 4 [3.5 to 5], miscellaneous 3.5 [3 to 5].
Discussion: Of the studies reporting the source of funding, RCTs funded solely or in part by for-profit organizations (n=18) favored the experimental therapy [median 3.75] more often than RCTs funded by not-for-profit organizations (n= 84) [median 3.5] (p=0.014). Similarly, the majority (59%) of RCTs funded by not-for-profit organizations had equipoise compared with only 8% of RCTs funded solely or in part by for-profit organizations (p=0.047). This study suggests that information on funding source may be important to the complete evaluation of published RCTs.
Objectives:
Methods: We focused on one disease category (multiple myeloma). We first sought all RCTs for this disease from 1996-1998, identifying 132 RCTs. To evaluate which treatment is better in a given RCT, we followed the intention of the original investigators and recorded them in the manner reported. The results of the each study were rated on a six point scale (1-standard treatment highly preferred, 6-experimental treatment highly preferred) with a value of 3.5 representing no difference between the two types of therapy (the point of equipoise). The features of RCTs which might influence reported results such as the type of treatment and the source of funding were studied. The rating of studies was performed by a single rater (ML) who was blinded to the source of funding. About 10-20% of all identified trials were also independently reviewed by the first author. Statistical inference on subgroups was performed utilizing the Mann-Whitney rank sum test and chi-square test for comparison of rankings and proportions, respectively
Results: Overall, the uncertainty principle was found to hold true in the conduct of RCTs in multiple myeloma. The median evaluation score for all 132 RCTs was 3.5 (63% of RCTs), ranging from a score of 2 (3%) to a score 6 (0.75%; only 1 RCT had score of 6). When the data were analyzed according to the type of treatment, the following scores (median [range]) were observed: induction 3.5 [2 to 5], maintenance 3.5 [3.5 to 5], supportive care 4.5 [3.5 to 6], salvage 3.5 [3.5 to 3.5], consolidation 4 [3.5 to 5], miscellaneous 3.5 [3 to 5].
Discussion: Of the studies reporting the source of funding, RCTs funded solely or in part by for-profit organizations (n=18) favored the experimental therapy [median 3.75] more often than RCTs funded by not-for-profit organizations (n= 84) [median 3.5] (p=0.014). Similarly, the majority (59%) of RCTs funded by not-for-profit organizations had equipoise compared with only 8% of RCTs funded solely or in part by for-profit organizations (p=0.047). This study suggests that information on funding source may be important to the complete evaluation of published RCTs.