Applied Swarm-based medicine: collecting decision trees for patterns of algorithms analysis

Abstract

Background: The objective consensus methodology has recently been applied in consensus finding in several studies on medical decision-making among clinical experts or guidelines. The main advantages of this method are an automated analysis and comparison of treatment algorithms of the participating centers which can be performed anonymously.

Methods: Based on the experience from completed consensus analyses, the main steps for the successful implementation of the objective consensus methodology were identified and discussed among the main investigators.

Results: The following steps for the successful collection and conversion of decision trees were identified and defined in detail: problem definition, population selection, draft input collection, tree conversion, criteria adaptation, problem re-evaluation, results distribution and refinement, tree finalisation, and analysis.

Conclusion: This manuscript provides information on the main steps for successful collection of decision trees and summarizes important aspects at each point of the analysis.

Keywords: Cancer; Consensus; Consensus finding; Decision tree; Radiotherapy; Swarm-based medicine.

Fig. 1

Workflow for patterns of algorithms analyses based on decision trees

Fig. 2

A sample draft decision tree, containing several imperfections. Text below the figure: (PC pancreatic cancer, BSC best supportive care, Gem gemcitabine)

Fig. 3

A sample of the first decision tree sent to the participating centre for validation and correction. Text below the figure: At the stage of refining the decision trees, any format can be used (in this example we used a decision tree in an email, which was discussed over phone, printed out and corrected with pencil). mCCRCC: metastatic clear cell renal cancer HI: hepatic insufficiency, SUN: sunitinib, PS: performance status, ZZ: Zugzwang, PAZ: pazopanib, HD IL-2: high-dose interleukin 2)

Fig. 4

A mode decision tree summarising and quantifying the radiotherapy recommendations of 24 Swiss centers for localised prostate cancer [15]. Text below the figure: The figure shows very low consensus for radiotherapy dose and duration of androgen deprivation therapy based on PSA value, T stage and Gleason score in prostate cancer. PSA = Prostate-specific antigen level (μg/L). Gl = Gleason. ADT = Androgen deprivation therapy

Fig. 5

Profile of criteria being used in decision making for recurrent glioblastoma, adapted from Hundsberger et al. [17]. Text below the figure: Criteria used by a anonymised centers in decision making displayed by green squares, others in red. MGMT = O6-methylguanin-DNA-methyltransferase promotor methylation status

Fig. 6

Treatment profile, the figures indicates the various treatments being offered in the context of first line metastatic renal cell cancer, adapted from Rothermundt et al. [21]. Text below the figure: Recommended treatment option from a specific center are displayed by green squares, others in red. The participating centres were named with their permission. INFα = interferon alpha; BEV = bevacizumab; SUN = sunitinib; PAZ = pazopanib; SOR = sorafenib; AXI = axitinib; EVE = everolimus; TEM = temserolimus; BSC = best supportive care