Affidea BV – Best Quality Data Model FINALIST – 2018
JURY’S REASONING
Affidea is a provider of diagnostic imaging and cancer care services in Europe. It owns and operates 235 medical centers across 16 countries with over 4400 employees, 870 of whom are medical doctors. Affidea serves more than 5.7M patients per year with revenue Eur400M.
In 2015 Affidea introduced the Dose Excellence Program for 72 types of CT scans in 12 countries. The system uses data from 75,000 examinations per month and analyses the dosage based on body type and CT gantry positioning. High dose levels are flagged by the system. This has led to Affidea creating 105 unified protocols to standardize operating procedures for CT scans. This program is key to Affidea’s strategy for winning international tenders.
Application
Briefly describe the organisation giving number of facilities, staff, revenue numbers.
Affidea is the leading provider of medical imaging services in Europe, operating 235 outpatient and PPP centers in 16 countries: 6.5 m patients, 1226 modalities, 7700 professionals, 2200 doctors. Revenue numbers: 2015 – 306,634 // 2016 – 310,954 (in € 000).
Please briefly describe the medical service, which the organisation is delivering in which it has deployed the best quality data model
Magnetic Resonance Imaging (MRI) is not only one of the most complex parts of the diagnostic chain, it is also one of the most expensive. Obtaining excellent anatomical and functional images is time consuming and requires high levels of technical expertise. Although vendors are attempting to codify workflows, there are numerous factors that lead to increasing heterogeneity in the way images are produced and the time it takes to produce them. The Affidea MR Excellence programm was developed to address these issues.
Please describe the way the organisation has deployed and used the best quality data model.
How has it changed the way it collects and manages data?
Our commitment is to continually improve and adapt in order to improve the safety and effectiveness of imaging care throughout our organization.
The Affidea MR Excellence Program (MREP) was established to address the balance between obtaining excellent anatomical and functional images, the high levels of expertise needed to perform MRI examinations, operational efficiency and maintaining diagnostic confidence and patient comfort.
MREP offers a unique opportunity to benefit from “collective optimization” efforts.
An international working group was formed, comprising six clinical radiologists led by an MR Biomedical Engineer with 25 years of experience on MRI protocol optimisation. By consensus, standardized protocols were developed in several workshops. For each indication and body part it was decided which sequences should be divided into which protocol, dividing them into four categories: i) core sequences, ii) recommended sequences, iii) conditional sequences, and iv) optional sequences. The optimisation part was conducted by a combination of site visits of the MR Applications Consultants and remote access to a Protocol Exchange Platform that was developed to easily transfer optimised protocols to those sites where no site visit was performed.
The Affidea Imaging Metrics (AIM) platform was developed as a web-based system to collect, process, quantify and present metrics retrieved from local ‘agents’ on an interactive dashboard. These agents transmit anonymised metadata derived from the DICOM headers through secure FTP to a central server. These data are compared with a reference database and compliance to standardized protocols calculated. In addition, deviation is defined as the percentage of sequences not included in the standardized protocols. Other key performance indicators include: number of exams, voxel size, examination time, and non-scanning time. After a period of monitoring, local teams assess workflows to reduce non-scanning time, using lean methodology and the AIM platform to assess impact. The user can setup his/her own alerts and when a specific KPI threshold (set by the user) is violated, an automatic e-mail is sent to the recipients that the user has defined.
How has that new quality data been used to change the way that health care services are delivered?
Affidea MR Excellence Program delivers the following outcomes, that are capable of change the way of health care services:Affidea MR Excellence Program delivers the following outcomes, that are capable of change the way of health care services:- The program significantly reduces the MRI examination times, 14.3% as per our last report. The time reduction has also positive effect on the patient experience, since the patient needs to spend less time in the MRI scanner bore, which is limited in space and could cause claustrophobia and panic.- The program also brings operational benefit due to increased number of daily examinations, throughput. 17.5% incensement on daily number of examinations, as per our last report.The Affidea MR Excellence Program (MREP) was established to address the balance between obtaining excellent anatomical and functional images, the high levels of expertise needed to perform MRI examinations, operational efficiency and maintaining diagnostic confidence and patient comfort.
When did the quality data model start affecting service delivery?
Month : July
Year : 2017
What are the main key performance indicators? How does the organisation measure the success of the project?
Definitions of Affidea MR Excellence Program KPIs:
NoE: Total Number of Exams performed within a user defined time interval.
On pace for is the projection to the end of the current month.
% change as compared to the average NoE value over an equal duration time interval in the past.
TAT: Total Acquisition Time is defined as the time point when the last sequence started – time point when the first sequence started + the duration of the last sequence.
Average TAT is calculated over a user defined interval.
Previous Period: is the average value calculated over an equal to the user defined time interval in the past.
% change as compared to the average NoE value over an equal duration time interval in the past.
Non Scanning Time: Non Scanning Time is calculated subtracting the time point of the first sequence of current exam from the time point of the last sequence of the previous exam minus the duration of the last sequence of the previous exam.
Average NST is calculated over a user defined interval.
Previous Period: is the average value calculated over an equal to the user defined time interval in the past.
% change as compared to the average NST value over an equal duration time interval in the past.
CMPL: Compliance to Standardized Protocols is calculated as the percentage of sequences that are included in the current exam and the standardized protocol.
Average CMPL is calculated over a user defined interval.
Previous Period: is the average value calculated over an equal to the user defined time interval in the past.
% change as compared to the average CMPL value over an equal duration time interval in the past.
DEV: Deviation is defined as the percentage of “foreign to the Standardized Protocol” sequences.
Average DEV is calculated over a user defined interval.
Previous Period: is the average value calculated over an equal to the user defined time interval in the past.
% change as compared to the average DEV value over an equal duration time interval in the past.
RES: Image Resolution is the average voxel size of all sequences within an exam in cc
Average RES is calculated over a user defined interval.
Previous Period: is the average value calculated over an equal to the user defined time interval in the past.
% change as compared to the average RES value over an equal duration time interval in the past.
Mapped Examinations are those exams that has been associated with reference protocols
Average Map value is calculated over a user defined interval.
% MAP: Percentage of recognized exams (mapped) over the total number of exams
Sequence Performance Index: To decide which sequences should be used on each site in terms of sequence parameters, based on the Sequence Performance Index.
SPI = (IQ*NoSl)/(ST*Res)
IQ: Input from Radiologist
NoSl: Number of slices
ST: Sequence Scan Time
RES: Sequence Voxel size