Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
Title: Single Patient Based Medicine can play a pivotal role in all clinical Trials.
- Sergio Stagnaro (2 July 2004)
-
Complications of NSAID therapy
- Colin Froman (3 July 2004)
-
Effect modification and generalisibility
- Sigrid M. Jansen, Sita M. Vermeulen, Patrick J. Bindels, and Gerben ter Riet. (28 July 2004)
-
Query re Table
- Neal Maskrey (20 December 2004)
|
|
|||
|
Sergio Stagnaro, Specialist in Blood, Gastrointestinal, and Metabolic Diseases Researcher in Biophysical Semeiotics Via Erasmo Piaggio 23/8 16037 Riva Trigoso (Genova ) Italy
Send response to journal:
|
Sirs, We must agree with the authors of such as interesting paper (1), who state correctly that “To provide safe and effective interventions for people, reliable and valid evidence is needed”. In doing that, of course, we need to undertake trials in samples of people who are as homogeneous as possible and applying the results to similar, well defined groups of patients. At this point, I underline the paramount role played by Single Patient Based Medicine, beside EBM (2, 3, 4, 5) (See HONCode website www.semeioticabiofisica.it). In a few words, it is unavoidable necessary to know personal inherited predispositions (constitutions) of each individual who will participate in clinical trials. In addition, doctor must perfectly evaluate his (her) single patient from his (her) singular constitutions, in order to obtain the most advantages for each patient from whatever clinical trials. This accounts for the reason that “Planning for the EU public Health Portal” all’URL: http://europa.eu.int/comm/health/ph_information/documents/ev_20030710_co01_en.pdf, suggests the SPBM as usefull tool in the primary prevention against malignancy: a 46-year-long clinical experience, in fact, allows me to state that without Oncological Terrain, cancer is not possible at all. (6) (See the above-cited website). 1) Dieppe P., Bartlett Ch., Davey P., Ebrahim S. Balancing benefits and harms: the example of non-steroidal anti-inflammatory drugs BMJ 2004;329:31-34 (3 July), doi:10.1136/bmj.329.7456.31 2) Stagnaro S. Single Patient Based Medicine against Health Care Disparities. (21 May 2004) http://bmj.bmjjournals.com/cgi/eletters/328/7450/1213 3) Stagnaro S. “Single Patient Based Medicine” versus EBM. (16 May 2003) http://bmj.com/cgi/eletters/326/7398/1048#32299 4) Stagnaro S. Single Patient Based Medicine against Health Care Disparities. (21 May 2004) http://bmj.bmjjournals.com/cgi/eletters/328/7450/1213 5) Stagnaro Sergio, Stagnaro-Neri Marina. SINGLE PATIENT BASED MEDICINE. Aspetti teorici e pratici della Medicina Basata sul Singolo Paziente. Ed. Travel Factory, Roma (in press). 6) Stagnaro-Neri Marina, Stagnaro Sergio. Introduzione alla Semeiotica Biofisica. Il Terreno oncologico”. Travel Factory SRL., Roma, 2004. http://www.travelfactory.it/semeiotica_biofisica.htm Competing interests: None declared |
|||
|
|
|||
|
Colin Froman, retired neurosurgeon home. 12 ramat Yam Herzlia 45861
Send response to journal:
|
The contribution reinforces the view that there are lies, damn lies and statistics. If 2-4 times as many users of NSAID present with renal pathology than non- users, those NSAID users also ingested analgesics in quantity. The worst complications of NSAID are not considered. Abscesses in fatty buttocks which were the repository of the standard GP Voltaren injection were not all that rare in my practice years. And there was a cardiologist, FRCP no less,who had severe premature cervical and lumbar disc disease,who had routinely given himself intravenous Voltaren. He couldn't understand why all his upper limb superficial veins had disappeared. He was lucky not to have dropped dead. The gastric tolerance of cox inhibitors has made for greater patient tolerance. My personal experience of profound relief of joint pain from these drugs and I for one would not forego their use on the grounds of the remote risks postulated by the article. Competing interests: None declared |
|||
|
|
|||
|
Sigrid M. Jansen, pharmacist Dept General Practice, Academic Medical Center, 1100 AZ Amsterdam, Netherlands, Sita M. Vermeulen, Patrick J. Bindels, and Gerben ter Riet.
Send response to journal:
|
Dear Editor, Dieppe et al.’s paper addresses an important issue, but poses
several difficulties1. In their 1st paragraph the authors state that ideally we
should undertake “trials in samples of people who are as homogeneous as
possible”. Note that the philosophy underlying the requirement of homogeneity
is that of scientific
generalisibility2. Briefly, in this philosophy generalisations are
guided by scientific insights, and not merely seen as “a mechanical aspect of
sampling”3. In the Results section they note that “those over 75
[years old] were excluded from most
[our italics] trials.” Here the underlying philosophy appears to be that of statistical generalisibility4.
We believe that the two philosophies are incompatible and that statistical generalisibility – although useful in survey-type research
– is an inadequate paradigm for scientific medical research. For example, when
we conservatively insert the value of 80% in Dieppe et al’s above-quoted
sentence, it would read “those over 75 [years old] were excluded from 80% [of
the 219] trials”. This would still imply that 20% of trials (that is over 40) did include >75-year-olds. It is
conceivable that the aggregate number of over-75-year-olds in those 40 trials
is sufficient to assess whether NSAIDS intended effects are modified by
advanced age. Of course, this may require the availability of individual
patient data, but that is another matter. Another objection against the
authors’ stratified sampling procedure, that left 194 eligible trials unassessed, is that a focus on the large trials (n ≥
200 per arm) would have been far more informative to the study of adverse
effects. It is ironic that the internal validity of this paper seems to have
suffered from the application of statistical generalisibility
that treats scientific problems as sampling problems too often. Conceptually, we believe that the issue that Dieppe et al. should have
addressed is effect modification, not
diversity. Their examples of diversity are restricted to the conventional set
of politically correct ones: “age, sex, gender[?], and
ethnicity”. Fundamentally, however, we think that the issue is about any
(combination of) factors – effect modifiers – capable of modifying treatment
effects, or susceptibility to adverse effects. Once that is agreed, it is clear
that each treatment-outcome relation may have its own set of potential
modifiers. Diversity therein is of course much larger than the received set of
sex, age and ethnicity. In summary, the challenge for trialists lies
in defining candidate determinants that may cause effect modification and in
ensuring sufficient representation of these in the study population. This will
generally imply over-inclusion of certain subgroups of patients instead of
representative sampling of a general patient population. Over time, complying
with this principle is likely to produce the knowledge needed for a truly
equitable health care in which claims of special effects in subgroups, whether
defined by age, sex, and/or ethnicity or otherwise, can be quantified and substantiated. References
Competing interests: None declared |
|||
|
|
|||
|
Neal Maskrey, Medical Director National Prescribing Centre, Liverpool L69 3GF
Send response to journal:
|
In this excellent paper, why are the values for unity for relative risk 0 in the table. Should they not be 1? Regards Neal Maskrey Competing interests: None declared |
|||