Business Analytics (BA) is a data management solution extensively used in non-medical business. BA refers to tools and software that help to collect, store, and analyze the data generated by a company’s activities by providing metrics and graphical dashboards to have an objective real-time measurement of business efficiency [1]. BA comprises data mining, process analysis, performance benchmarking, and descriptive analytics, providing a complete overview of the situation in real time [1,2]. BA has three main branches: 1) descriptive analytics, a summary of current and previous data in tables, graphics and reports; 2) predictive analytics, a tool that analyzes present and past efficiency data, and 3) prescriptive analytics, to identify the best alternative in the management of a particular scenario using descriptive analytics. The BA approach is gaining a pivotal role in decision-making processes, and big data represent an added value to improve efficiency and performance.
The implementation of BA in medicine, and in radiology in particular, is a slow process, with only a few scattered experiences throughout the world [3-5]. Radiology, compared with other disciplines, has the advantage of having a fully computerized data archive resulting in the possibility of tracking workflows and gaining insight into a department’s productivity and efficiency [4]. Databases, dashboards, charts, and graphics provide a detailed real-time picture of the situation and are helpful for business administration, monitoring productivity and making nonarbitrary decisions.
The Emergency Radiology Department is an area where the implementation of BA can help to improve efficiency – by monitoring workloads, scanner occupation, and turnaround times (TATs) – and, at the same time, to prevent radiologist burn-out.
Since March 2019, a project for the implementation of a BA system with descriptive features to support management decisions at the Radiology Department of Sant’Andrea University Hospital in Rome, Italy, was started. The process was rather complex and, apart from the endorsement of the Hospital’s General and Healthcare Directorates, required the collaboration of many professional figures, including radiologists, information technologists and clinical engineers, accountants from the management control office, and industry partners. The first step was to analyze the data warehouse and to create connections among the different data repositories. Moreover, an exchange of information with the regional software for emergencies was implemented to monitor the activity of the Emergency Department. The second step was to identify key performance indicators and to create the dashboards necessary to monitor the different activities. Each dashboard contains specific filters that allow the immediate extraction of data, which are constantly updated at midnight of the previous day, that is, less than 24 hours behind the day on which the data are accessed. The third step, still ongoing, is the continuous data curation and data stewardship necessary to keep the data as clean as possible.
A specific dashboard for Emergency Radiology was created. This dashboard offers a real-time at-a-glance view of the number of accesses and examinations, the distribution of examinations per technique (X-ray, ultrasound, and CT), the hourly occupation of the scanners and the TATs. The TATs were segmented to monitor the different phases of the process: “global” TAT (from examination order to final reporting); “execution” TAT (from examination order to examination execution); “reporting” TAT (from the end of the examination to final reporting). Radiologist workload is monitored through dashboards analyzing the overall number of reports, weight of different examinations performed by the single radiologist, mean and median reporting time, and distribution of reporting per color-coding (i.e. number of “blue-codes” reported per radiologist).
The most important test to verify the usefulness of BA in the management of an Emergency Radiology Department was provided by the COVID-19 pandemic. The COVID-19 pandemic had overwhelming effects on healthcare systems worldwide due to the huge number of patients accessing the emergency rooms; this pressure brought some hospitals to the brink of collapse and their personnel to burn out. At the same time, hospitals were unable to guarantee continuity of care for fragile and cancer patients.
In our personal experience, BA software allowed us to understand in real-time changes in workflows related to the pandemic and to react accordingly. We realized that the complete shut-down of the Radiology Department at the time of the first pandemic wave (March-April 2020) – when human resources were re-allocated to guarantee specific emergency radiology pathways for positive COVID-19 patients and outpatient services were either reduced or completely closed – had been driven by emotional decisions, not supported by actual data on the number of patients accesses, scanner occupation, and radiologist workloads. For this reason, during the second pandemic wave, which occurred in Rome in November 2020, it was decided to maintain two parallel pathways for COVID-19 and non-COVID-19 patients, avoiding a complete shut-down of the Radiology Department. In this way, despite the 21.5% reduction of total imaging volume due to the COVID-19 pandemic in 2020 compared to 2019, management decisions supported by dedicated BA software made it possible to increase the number of CT examinations in fragile/oncologic outpatients by 11.7%. More importantly, this result was obtained without significantly affecting emergency radiology TATs or radiologist workloads [5].
In conclusion, resilience is fundamental during crises because it enables an immediate positive reaction. BA is an extremely powerful tool that provides precise real-time insight into changing scenarios and, when applied to our Radiology Department, it allowed us to make timely evidence-based organizational decisions, minimizing loss of productivity and maximizing efficiency as well as overall welfare, without negatively affecting radiologist workloads.
References
1. Cook TS, Nagy P. Business intelligence for the radiologist: making your data work for you. J Am Coll Radiol. 2014 Dec;11(12 Pt B):1238-1240.
2. Nielsen S. The impact of Business Analytics on management accounting. SSRN Electronic Journal, 2015. Available at SSRN: https://ssrn.com/abstract=2616363 or http://dx.doi.org/10.2139/ssrn.2616363
3. Martin-Carreras T, Chen P-H. From data to value: How Artificial Intelligence augments the radiology business to create value. Semin Musculoskelet Radiol. 2020 Feb;24(1):65-73.
4. Sigler R, Morrison J, Moriarity AK. The Importance of Data Analytics and Business Intelligence for radiologists. J Am Coll Radiol. 2020 Apr;17(4):511-514.
5. Laghi A, Tamburi V, Polici M et al. Management decisions of an Academic Radiology Department during COVID-19 pandemic: the important support of a business analytics software. Eur Radiol. 2022 Oct;32(10):7048-7055.