Alert Management - How to Find the Needle in the Haystack
Identifying Cluster of “False Positives” via Smart Root Cause Analysis
Currently, many alerts are being raised in the European Medicines Verification System, creating a high workload for the different stakeholders. Most of them are “false positives” due to user errors or technical errors like misconfigured scanners. As a consequence of the European Delegated Regulation 2016/161/EU, a European Medicines Verification System (NMVS) has been implemented in order to avoid that falsified medicines can enter the supply chain and be distributed to the patient.
Today, there are more than 1.000 pharmaceutical companies (Onboarding partners - OBP) are uploading pack data to the system, more than 100.000 end users connected and more than 200 million scans performed every week.
The Problem of Having Too Many Alerts in the System
As a safety feature, each pack covered by this legislation is equipped with a unique identifier containing a product code, batch number, expiry date, and serial number. These unique identifiers are stored in a National Medicines Verification System. At the moment of dispense, the pack is scanned, and its authenticity is verified.
If the data that has been scanned do not coincide with the data stored in the NMVS, an alert is raised. Currently, there are many alerts being raised, most of them “false positives” that are based on technical errors or user errors. This is creating a huge “haystack” of alerts and make it hard to identify the real falsifications. To investigate an alert consume a lot of resources and creates a huge workload for all stakeholders.
Where Do Alerts Come From?
Alerts are being generated either due to a mistake on the OBP side or the end-user side. Both technical and manual errors can create alerts. Frequent root causes are that a product has been put into the market, and the corresponding information was not uploaded.
Also, it might happen that coding standards are not being fulfilled. The end-user has the possibility to read a code with a scanner. If the scanner is not correctly configured, an alert is raised. Also, a pack can be deactivated by entering the unique identifier manually which might lead to typos.
How Can Alert Management Tools Help to Reduce Alerts?
To investigate the root cause of an alert, different stakeholders must exchange information regarding the affected pack. Therefore the European Medicines Verification Organisation (EMVO) is setting up an Alert Management System (AMS) consisting of different components: an AMS HUB, AMS for OBPs, and also a National Alert Management System, allowing the various stakeholders (end users, NMVOs, OBPs, etc.) to investigate the root cause of the alert.
Today, more than 300.000 alerts need to be investigated every week, and a single investigation can take several hours. The analysis and handling of alerts are often done manually. Alert management systems - like the one from Arvato Systems - therefore enable the following functions:
Alert Management Tool Functions
The overall goal is to reduce the number of alerts in the system and thus the resources required for alert investigation. It also includes the following functions:
An alert management tool needs to receive alerts and give the user the possibility to manage them effectively. Arvato’s alert management tools visualize alerts in a user-friendly way and allow the user to filter, order, and classify them. During the investigation, an alert can be commented on, a status can be set, and it can be assigned to someone. Dashboards give a quick overview of the main root causes, how alerts evolved over time etc.
Identify the Root Cause
Using big data and Artificial Intelligence tools, it is possible to identify the root cause of the vast majority of the alerts. Once the root cause is identified, the alerts can be clustered. Arvato Systems’ root cause analysis tool aims to identify the root cause and automatically classify it in 95% of the alerts. Like this, the user can focus on the investigation of the relevant alerts.
Communicate between the Different Stakeholders
Frequently the input from several users is needed in order to investigate an alert, therefore an efficient communication between the different parties is key. The ecosystem allows end-users, NMVO’s and pharmaceutical companies to interact in a secure and anonymous way. Also, documents and images can be shared. Predefined messages help to allow communication between stakeholders in different countries.
Would You like to Know More about the Alert Management System from Arvato Systems?
Then write to us! Our experts will be happy to get back to you and answer your individual questions.