gms | German Medical Science

11. Jahrestagung 2004 der GAA

Gesellschaft für Arzneimittelforschung und Arzneimittelepidemiologie

30.09. bis 01.10.2004, Jena

Innovative Methodology to Translate Conditions of Daily Practice into an Evidence Based Study Design: Example of a randomized feasibility trial to compare a world where inhaled insulin (INH) is available with a world without INH

Meeting Abstract

Search Medline for

  • corresponding author presenting/speaker E. Stridde - Pfizer GmbH/Karlsruhe
  • E. Huppertz - Aventis Pharma Deutschland GmbH/Bad Soden
  • L. Heinemann - Profil GmbH/Neuss

Gesellschaft für Arzneimittelanwendungsforschung und Arzneimittelepidemiologie (GAA) e.V.. 11. Jahrestagung der Gesellschaft für Arzneimittelforschung und Arzneimittelepidemiologie (GAA) e.V.. Jena, 30.09.-01.10.2004. Düsseldorf, Köln: German Medical Science; 2004. Doc04gaa22

The electronic version of this article is the complete one and can be found online at:

Published: September 30, 2004

© 2004 Stridde et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License ( You are free: to Share – to copy, distribute and transmit the work, provided the original author and source are credited.




„Conventional" clinical trials are required to obtain regulatory approval for new medication. However, in these „head to head" clinical trials new medication shows often the same efficacy as standard treatment but may have beneficial effects under conditions of daily practice. So daily practice should be translated into clinical trials. Alternative study designs are needed. A clinical trial to compare a „world" where a new medication is available with a „world" without this option could help to show beneficial effects.

Aim of the Study

1. Create a evident study design to show beneficial effects of a new form of application in comparison to standard treatment under conditions of daily practice.

2. Test the design in a pre-study (feasibility trial)

3. Estimate the proportion of patients who prefer this novel treatment options in comparison to standard therapy (needed for clinical trial sample size calculation).


T2Ds failing OAD were randomly (patient based) assigned to 2 groups (A / B). Both groups received the same educational material regarding the available treatment options. In addition group A received information on INH. The decision regarding future treatment was made during an interactive discussion of patients and physicians.


Proportions were analyzed using Fischer's exact test, and described using the odds ratio and theoretically exact 95% confidence intervals (CIs).


779 patients (Group A = 391, Group B = 388) were recruited. 169 patients (43.2%) in group A opted for a treatment that included insulin, in comparison with 60 patients (15.5%) in group B; odds ratio 4.16 (95% CI 2.93 to 5.95; p < 0.0001). The proportions, together with published results of the different treatment options, permit sample size calculation. The limitation of the study is that the decision was only theoretical. There was no drug intervention.


The described method is suitable for translating conditions of daily practice into an evidence based study design (randomization). The results require confirmation in a large clinical trial including drug intervention. The study design seems to be suitable to show beneficial effects of new medication, not seen in conventional clinical trials, also in other therapeutic areas.

Conflict of interest: Sponsor of the Study were Pfizer Inc./USA and Aventis Inc./USA.