Veterinary Medicine - September 2008 - (Page 484)

Leading Off Poor study design Disconcertingly, poor study design often weakens the validity of conclusions drawn from veterinary clinical trials. Many clinical trials lack appropriate controls, randomization of treatment groups, or blinding procedures. Statistical support is critical in well-designed trials but is often not used or is reserved until data have been collected. A common sequela of insufﬁcient or late statistical input is that only a small number of patients are studied, which leads to two common errors. • First, investigators may use the same animal to study multiple therapeutic interventions. Thus, even if randomization and crossover are properly performed, the increased number of comparisons increases the risk that a difference will be falsely identiﬁed as signiﬁcant. • Second, a small sample size decreases the power of the study to detect true treatment differences. The investigator, and practitioner, often confound the mistake by interpreting the lack of signiﬁcant difference as evidence that the treatment groups are the same, and thus, the treatments are equivalent—or in the case of a placebo-controlled study that the treatment had no effect. It is a veterinarian’s responsibility to learn what constitutes a well-designed clinical trial and to be able to assess the validity of the investigators’ conclusions. information. Frequently, only those reports in which their product performs well make it to peer-reviewed literature. Reports of studies sponsored by manufacturers should be considered biased but not necessarily suspect. Need for greater practitioner participation Although it is the investigator’s responsibility to design clinical trials that provide valid conclusions, often the best design is circumvented by lack of practitioner participation. However, as clinical research continues to struggle for acceptance in academia, the role of practitioners becomes increasingly important. Practitioners have historically provided the patient base for industry-based clinical trials, and private practitioners are best-positioned to assume a leading role in implementing clinical research. As a clinical pharmacologist, I have concentrated on clinical research. Our laboratory has implemented a number of clinical trials, often in partnership with private practices. I am humbled by the care and attention to detail that participating practitioners and their clients dedicate to our studies. However, some participants fail to follow study protocols, not realizing that doing so excludes that patient’s data from the study and reduces the sample size. Furthermore, we have difﬁculty identifying potential participating practitioners and clients. For example, we have three ongoing clinical trials, two that involve compounded preparations (including transdermal gels) and one that investigates antimicrobial resistance. Practitioners enthusiastically agree that the medical information to be generated by the successful implementation of these studies will meet practitioner and patient needs, yet we have not identiﬁed sufﬁcient participants. Practitioners express several con- Transparency of potential bias Identifying bias is also important. Readers should be aware of a study’s source of funding. As mandated by the Food and Drug Administration (FDA), manufacturer-sponsored studies that support drug approval generally test hypotheses that the manufacturer’s product is not inferior to a positive control, rather than generate unique 484 September 2008 VETERINARY MEDICINE

Table of Contents Feed for the Digital Edition of Veterinary Medicine - September 2008

Veterinary Medicine - September 2008 Contents Leading Off Practical Matters Idea Exchange A Challengin Case: Thymic Cyst and Recurrent Chylothorax in a Cat 10 Life-Threatening Behavior Myths CE Form Advertiser Index Marketplace/Classifieds Mind Over Miller

Veterinary Medicine - September 2008

For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.