Systematic Reviews / Meta-Analysis
Johannes Vey, M.Sc. (Head of the working group)
Prof. Dr. Meinhard Kieser
Junior Sinclair Awounvo Awounvo, M.Sc.
Amina Cheikh Rouhou, M.Sc.
Marie Merling, M.Sc.
Dr. Tanja Proctor
Samuel Zimmermann, M.Res
The working group "Systematic Reviews / Meta-Analysis" applies and develops statistical methods for evidence synthesis.
Systematic reviews provide an overview over a specific (medical) topic by systematically identifying and compiling results of multiple different research studies. They thus enable medical decision-making along the principles of evidence-based medicine.
Meta-analysis is a quantitative method that allows statistical synthesis of such research results. Applications of meta-analyses are highly diverse and range from the efficacy of specific (medical) interventions such as surgical procedures, drug therapies or diet changes, to diagnostic tools and prognostic models.
The research interest of our working group lies in methods for indirect comparisons, meta-analyses with biomarker subgroups and network meta-analyses with sparse networks. Research projects are often collaborations with other institutes on and off campus.
Research interests:
- Meta-Analysis and indirect comparisons for biomarker subgroups
- Methods for indirect comparisons
- Network meta-analysis in sparse networks
Examples for projects in the working group Systematic Reviews / Meta-Analysis
Cooperation with the Network Aging Research (NAR)
Participation in two Systematic Reviews of the Junior Research Group „Effects of physical activity and exercise on motor and cognitive performance in older adults“ with Dr. Schwenk. These systematic reviews and meta-analyses investigate two aspects of healthy aging. The Systematic Review (SR) EArly deals with prevention of infirmity that appears halfway through adult life. For this purpose, various sports and dietary interventions in this population group are searched, systematically processed and statistically summarized. The SR ENSuRE is concerned with whether intervention programs implemented through dietary changes or a specific nutritional program with elderly patients can slow down sarcopenia in old age.
- SR EArly (funding through BMBF 2020-2021) on the effect of physical activity and strength, balance ability, and falls in healthy adults aged 40-65 years: a systematic review and meta-analysis.
- SR ENSuRE (funding through BMBF 2017-2019) to determine the effect of physical training and nutrition interventions on sarcopenia-associated outcomes.
Link: https://www.nar.uni-heidelberg.de/juniorforscher/ng_schwenk/index.html
Surgical Reviews
Participation in Systematic Reviews of the Departement of Surgery of the University Medical Center Heidelberg, including systematic reviews regarding laproscopic and robot-assisted surgery.
The use of robot-assisted surgery has made high precision operations in confined spaces possible. This development concerns interventions that can be performed in a minimally invasive manner. While the use of more complex robot-assisted surgery is not necessary for all kinds of surgery, it excels in complicated cases where improved mobility and visualization are crucial.
- Ongoing: SR Robotic (Dr. Schmidt): The content of this systematic investigation is the question of transferability of surgical skills between different training and surgical methods. One object of investigation is whether surgical skills learned using virtual reality simulators can be transferred to the operating room.
Schmidt M. W., Köppinger K.F., Fan C., Kowalewski K.-F., Schmidt L. P., Vey J., Proctor T., Probst P., Bintintan V. V., Müller-Stich B.-P., Nickel F. (2021): From Virtual Reality to the Operating Room in Robotic-assisted Surgery – A Systematic Review and Meta-Analysis of Skill Transfer and Predictability of Skill. BJS Open (akzeptiert) - Ongoing: SR Robotic II (Dr. Schmidt): A follow-up review will now investigate whether general surgical and/or laparoscopic skills and experience contribute to faster learning and better performance of the robotic-assisted surgical procedure. The goal of this systematic review is to make training for robotic-assisted surgery as efficient as possible.
- Finalized: SR ROCobotic (Dr. Kowalewski): This systematic review compared functional outcomes in patients with colorectal cancer who underwent either laproscopic or robotic-assisted partial rectal resection surgery.
Kowalewski K. F., Seifert L., Ali S., Schmidt M. W., Seide S., Haney C., Tapking C., Shamiyeh A., Kulu Y., Hackert T., MüllerStich B. P., Nickel F. (2021): Functional outcomes after laparoscopic versus robotic-assisted rectal resection: A systematic review and meta-analysis. Surgical Endoscopy,35/1:81-95,2021
In cooperation with the working group "Systematic Reviews" of the Study Center of the German Society of Surgery (SDGC), we offer methodological consultations on statistical aspects of surgical reviews.
External Cooperation
Projects in cooperation with Dr. Ronellenfitsch, University of Halle
- Ongoing: A systematic review will consider preoperative chemoradiotherapy and preoperative chemotherapy compared with surgery without adjuvant therapy in adenocarcinoma of the esophagus and esophagogastric junction. These three treatment options will be analyzed with regard to their therapeutic success using individual patient data as well as aggregated data together in a network.
- Finalized: disease-free survival as a surrogate for overall survival in gastroesophageal adenocarcinomas. Aggregate data and individual patient data were aggregated for validation
Ronellenfitsch U, Jensen K, Seide S, Kieser M, Schwarzbach M, Slanger TE, Burmeister B, Kelsen D, Niedzwiecki D, Piessen G, Schuhmacher C, Urba S, van de Velde C, Ychou M, Hofheinz R, Lorenzen S, Disease-free survival as a surrogate for overall survival in neoadjuvant trials of gastroesophageal adenocarcinoma: Pooled analysis of individual patient data from randomised controlled trials, European Journal of Cancer, Volume 123, 2019, Pages 101-111
Ongoing methodical research:
- Investigation into the behavior of statistical methods of network meta-analysis in sparsely populated networks. This involves evaluating data-generating mechanisms, conducting a simulation study of different types of networks and methods, and developing a new way to visualize analysis results from network meta-analyses.
Seide S. E., Jensen K., Kieser M. (2020): A comparison of Bayesian and frequentist methods in random-effects network metaanalysis of binary data. Research Synthesis Methods,11/3:363-378, 2020
Seide S. E., Jensen K., Kieser M. (2021): Utilizing radar graphs in the visualization of simulation and estimation results in network meta-analysis. Research Synthesis Methods,12/1:96-105,2021 - Investigation into the possibility of combining biomarker-positive subgroups with evidence from the overall population via indirect comparisons and the use of aggregate data and individual patient data to improve estimation. Several Bayesian methods were explored in a simulation study.
Proctor T., Jensen K., Kieser M. (2020): Integrated evaluation of targeted and non-targeted therapies in a network meta-analysis. Biometrical Journal,62/3:777-789,2020 - Investigation into the possibility of using matching techniques to improve comparability between aggregate data and individual patient data. In two simulation studies, different practice-relevant scenarios were combined and investigated.
Weber D, Jensen K, Kieser M. Comparison of Methods for Estimating Therapy Effects by Indirect Comparisons: A Simulation Study. Medical Decision Making. 2020;40(5):644-654.