Stefanie Hennig (1), Charlotte Kloft (2)
1. University of Queensland, Brisbane, Australia; Alexander von Humboldt Foundation, Germany 2. Insitute fuer Pharmazie,Freie Universitaet Berlin, Berlin, Germany
Objectives: Pharmacometrics aims to understand the drug-patient interaction, describe the quantitative aspects of disease and pharmacology, connecting various fields such as physiology, pharmacology, pharmacotherapy, clinical pharmacy, mathematical modelling, statistics, systems biology, pharmacokinetics/-dynamics in a coherent framework to understand and improve human health. As such we need to communicate within multidisciplinary teams and convey our results to clinicians, editors and statisticians in a translational manner.
The aim of this review was to identify publications that have applied the nonlinear mixed effects (NLME) modelling approach since its first appearance in 1979, and have been published in high impact clinical journals.
Methods: The search terms “nonlinear mixed effect modeling” OR “nonlinear mixed effect modelling” OR nonmem OR monolix OR pharmacometric* were used to search three databases (Pubmed, Web of Science and Embase) for articles published in English between 1979 and 2018. Journal impact factor (IF) were identified via Web of Science or the journal’s webpage. Articles published in journals with an IF > 6.7 were read in full to examine the presentation of the NMLE methods. This cut-off was selected based on: i) CPT was the journal with the highest IF in the top 20 journals that NLME article are published in, ii) the IF of Clinical Pharmacology and Therapeutics (CPT) currently is 6.655, and iii) CPT holds position 7 when all journals were ranked based on the frequency of publishing NLME articles.
Results: Of the over 12,000 identified articles, 4910 articles remained after duplication were removed and titles and abstracts were scanned. An increasing number of articles per year were published since 1979 in 655 unique journals. A majority (51%) of these articles are published in only 13 journals. Journal’s impact factors (IF) ranged from 0.1 to 26.3 (Journal of Clinical Oncology). The journal (CPT: Pharmacometrics & Systems Pharmacology) most commonly publishing NLME articles has currently no IF. The median impact factor was 3.08. In total, 10.6% of articles are published in journals with no IF.
One-hundred (2.0%) articles are published in journals with IF> 6.7. Over 64.9% of these articles had a first/corresponding author from academia, 27.7% from a hospital and 7.4% from industry. The most common author teams were academics and clinicians (38.1%) followed by academics and authors from industry (14.9%). When published in high IF journals, presentation of the methods used and the description of the results was seldom according to standards, only 14% of article can be considered acceptable. [1, 2] Nineteen (19%) of the articles in higher IF journals have had a higher average number of citations per year compared to the journal’s IF, which is less compared to other fields. [3]
Conclusions: Communication of results and their impact arising from NLME studies could improve and ultimately patients’ outcome and health. Increased quality peer-review within the area and further engagement with statistician and clinicians may increase the opportunity of pharmacometrics to increase its impact.
References:
[1] Jamsen KM, McLeay SC, Barras MA, Green B. Reporting a population pharmacokinetic-pharmacodynamic study: a journal’s perspective. Clin Pharmacokinet. 2014;53(2):111-22.
[2] European Medicines Agency, Committee for medicinal products for human use. Guidline on reporting the results of population pharmacokinetic analyses. CHMP/EWP/185990/06. London 2007.
[3] Callaway E. Beat it, impact factor! Publishing elite turns against controversial metric. Nature. 2016;535(7611):210-1.
Reference: PAGE 28 (2019) Abstr 9234 [www.page-meeting.org/?abstract=9234]
Poster: Methodology - Other topics