A workflow management system for the biological samples exchange within the amyloidosis network

To allow a better coordination of the Italian amyloidosis network, and also to provide useful information for patients, some years ago a web portal (www.amiloidosi.it) was developed. One of its sections is devoted to the exchange of biological samples between peripheral institutions and the referral center. However, the section is purely descriptive, and does not support data collection and retrieval, or the possible problems that may occur along with the sample exchange process. The current workflow technology seems to offer a convenient solution to build a cooperative system for this kind of process and this article presents the design of a web-based workflow management system for the biological sample exchange. The final aim is to integrate the system within the portal. Introduction: Biological sample exchange is currently a high interest topic, as well as biobanking diffusion; there is a need for promoting networking of biobanks and for harmonizing the exchange procedures [1]. Other than biobank networks, the management of rare diseases, particularly in the context of clinical trials, often requires highly specialized tests, whose execution is not possible in every laboratory. Thus, in selected cases, peripheral network institutions (PNI) need to send biological samples to a referral coordination centre (RC) to perform tests. In the management of patients with amyloidosis, tests such as quantification of serum NT-proBNP, free light chains and serum amyloid A concentrations, and high-resolution serum/urine immunofixation, currently must be carried out at the IRCCS Policlinico San Matteo. More specifically, there are three different cases: (a) samples are sent to RC for the purposes of a clinical trial; (b) the RC asks for a sample from a PNI to solve a mismatch between test results performed in different settings; and (c) the PNI itself asks RC to perform a special test. In these cases, aspects like the scheduling of sample exchange, sample handling and shipment, and the proper communication of patient data and results are very critical. In fact, first of all the sample shipment must be agreed upon between the requesting and test-performing centers. Once the date has been arranged, samples must be sent by courier and must arrive within 24 h from the shipment. Samples must be associated with the necessary documentation that currently can be sent together with the sample or by regular mail, e-mail, fax, etc. This raises the issue of accuracy in matching the sample with its documentation. Currently, within the Italian Amyloidosis network, the whole process is managed by PNIs and RC through telephone calls, and a number of controls (about dispatch modality, date, time of arrival, etc.) are made manually. This worklow is time-consuming and poorly controlled. Therefore, we propose the implementation of a web-based workflow management system (WfMS) for managing the entire process. Methods: In the past, we have extensively exploited workflow technology in clinical domains [2–4]. Briefly, according to the Workflow Management Coalition, a WfMS is a ‘computerized system for process automation, that, running one or more Wf engines, executes a number of tasks, based on a computer representation of the logical process’ [5]. The WfMS focus is on organizational aspects of work with regard to a possible support of its execution through information and communication technology. Wfs are case-based; an instance represents a single enactment of the process, during which information, documents and tasks are passed from one participant to another. A participant is a resource, either human or non-human, involved in the process, which performs the work described in the process definition. In this study, an instance represents the execution of the workflow for a particular test request, and the resources involved are doctors, lab technicians, laboratory instrumentation, and biological samples. The WfMS implements a finite-state machine for the request, starting from the initial centre demand, through the sending of biological sample, the analysis, and its reporting. We implemented the WfMS using Oracle Workflow and the workflow process has been realized within the graphical interface of Oracle Workflow Builder (see a portion of the process in Figure 1). The system proposed is web-based, because we intend to use the interface offered by the amyloidosis portal. The WfMS accesses the patients’ data stored through the amyloidosis portal and also a specifically developed database containing all the relevant information about laboratory tests and analysis requirements. The database also allows the storage of all information associated with the requests and the samples exchanged. Results and discussion: A Wf model has been developed which allows the implementation of the following process steps, each one corresponding to specific states of the request: 233