Deceit by patients can be generalized in clinical trials, researchers warn. The degree of cheating in a test - a whopping 30% - is "surprisingly high" and "worrying", two teams report this month Chest , the journal of the American College of Chest Physicians. Based on this and other recent studies, the authors warn that "some kind of duplicity ... can be expected in almost all clinical trials."
The massive cheating turned after two research teams - one at Johns Hopkins University and the other at the University of California, Los Angeles (UCLA) - examined how often former smokers suffering from obstructive lung disease are unable to use the bronchodilator inhalers. Michael Simmons, a UCLA biostatistician, and four colleagues told volunteers in this trial that the use of drugs would be monitored. But they do not mention that a small device on the inhaler was recording the exact time and extent of use.
When the researchers looked at patterns of use of the inhaler later, they found that many patients just poured the inhalers before their quarterly clinic visits - indicating that they dumping of the drug in order to appear to be in compliance with the treatment regimen. Among 101 patients monitored for a year, 30 "undervalued" drug from an inhaler at least once misrepresented a doctor during an examination that they had taken the drug.
Cynthia Rand, a health psychologist at Hopkins and co-author of the study, explains that many other researchers identified the non-compliance as a problem, but this study breaks all previous records for the extent of the deception. To the surprise of researchers, patients duplicity proved to be "our finest, eager patients to please, the most enjoyable," said Rand. After reviewing dozens of factors such as race, gender, education, smoking history, the researchers were unable to find the quality that set apart cheaters. In addition, those who were dishonest about taking their medication were less likely to lie about current smoking (verified by testing saliva).
Rand said there is no way to predict who will cheat. An important lesson, she said, is that no amount of exhortation will make consistent volunteers -. Unless supported by a good method of checking
cheating patient may have "serious and potentially significant" impact on the results of clinical research, said Rand. If unsupported, cheaters are misclassified as followers scrupulous, introducing "a lot of noise" in the data. Analysis of dose-response calculations can be badly skewed in studies that do not carefully monitor cheating, she fears.
The extent of cheating in this study is a revelation, says psychiatrist Bruce Bender of the National Jewish Medical and Research Center in Denver, Colorado. The report is not good news, Bender adds, because it suggests that many clinical trials may be underestimating the potency of the drug.
Related Sites
Summary Chest article
0 Komentar