Pearn en Chalmers are interested to hear whether researchers in the Netherlands share Vandenbroucke's position that it is not unethical to selectively report well-designed research (). As a clinical pharmacologist involved in research projects based on solid contracts with funding bodies and individual patients, I have to endorse Pearn and Chalmer's position and strongly disagree with Vandenbroucke as I did three years ago when the latter first expressed his views. Withholding results of well-designed research from publication is increasingly being practiced in industry-funded research which in the current context of scarcity is the mainstay of clinical research worldwide. This is particularly so when the results do not fit in the company's selling policy. This problem is far more serious than what Vandenbroucke calls the silent burial of unspectacular data. Investigators agreeing on the Declaration of Helsinki have a commitment to the patients who were compliant with placebo treatment in spite of illness, and to the community at large to communicate their knowledge even if such knowledge represents unwelcome thruths to the funders.
A special case in this context can be made for negative trials. They can be either truly or erroneously negative. Erroneously negative means that we have to accept the null hypothesis of no treatment difference but that we do so with far too little power. For example, accepting the null hypothesis with p = 0,15, n = 30-60, means that we have about 30% chance that a true difference is missed. Unfortunately, many negative studies fall into this category. Not to publish such work seems to be defendable because of its limited value and the potential harm of wrong conclusions that might result from such studies. The position of Vandenbroucke not to publish even statistically powerful research routinely, although such an approach served the shifting of unimportant data in the past, is no longer warranted in the current field of research based upon agreed multicenter protocols and informed consent. Osler and Harvey who were both very convinced of physicians having a mission, would have agreed if in their time there had been so much high-quality research.
Is selective reporting of well-designed clinical research unethical as well as unscientific?
Utrecht, january 1996,
Pearn en Chalmers are interested to hear whether researchers in the Netherlands share Vandenbroucke's position that it is not unethical to selectively report well-designed research (). As a clinical pharmacologist involved in research projects based on solid contracts with funding bodies and individual patients, I have to endorse Pearn and Chalmer's position and strongly disagree with Vandenbroucke as I did three years ago when the latter first expressed his views. Withholding results of well-designed research from publication is increasingly being practiced in industry-funded research which in the current context of scarcity is the mainstay of clinical research worldwide. This is particularly so when the results do not fit in the company's selling policy. This problem is far more serious than what Vandenbroucke calls the silent burial of unspectacular data. Investigators agreeing on the Declaration of Helsinki have a commitment to the patients who were compliant with placebo treatment in spite of illness, and to the community at large to communicate their knowledge even if such knowledge represents unwelcome thruths to the funders.
A special case in this context can be made for negative trials. They can be either truly or erroneously negative. Erroneously negative means that we have to accept the null hypothesis of no treatment difference but that we do so with far too little power. For example, accepting the null hypothesis with p = 0,15, n = 30-60, means that we have about 30% chance that a true difference is missed. Unfortunately, many negative studies fall into this category. Not to publish such work seems to be defendable because of its limited value and the potential harm of wrong conclusions that might result from such studies. The position of Vandenbroucke not to publish even statistically powerful research routinely, although such an approach served the shifting of unimportant data in the past, is no longer warranted in the current field of research based upon agreed multicenter protocols and informed consent. Osler and Harvey who were both very convinced of physicians having a mission, would have agreed if in their time there had been so much high-quality research.