Novel model for automatized insight into quality of care without additional administrative burden
02 March 2018
Author: Rolf Bremmer
AMSTERDAM: LOGEX, a Dutch healthcare analytics firm, in cooperation with local gastroenterologists, has developed a new model for providing automatized short-term clinical outcomes per intervention. This big data approach, based on hospital information system data will be published this week in BMJ Open – the open access label of British Medical Journal. The published methodology, Textbook Outcome, is promising, since it can be applied to a large number of high-volume medical interventions without any additional registration. This publication marks an important step towards outcome measurement of healthcare for medical specialists – and ultimately for patients – especially suited for interventions where traditional methods and extensive clinical registration are not feasible.
Today, both patients and health care providers require more transparency of and insight into the quality of care, yet it is undesirable to burden doctors with additional administrative tasks. Using the novel approach – Textbook Outcome – we demonstrate for the first time insight into short term outcomes, without additional administration.
Textbook Outcome is a composite outcome measure that considers whether a patient has received an optimal treatment process without any adverse events: no re-interventions, and excess inpatient stay and no return to the emergency room. The novelty of this approach is that it can be determined on the basis of ready-available hospital information system data which is necessary for reimbursement from health insurance companies. Until now, this data is not used for providing insight into the quality of care. Yet, by remodeling this data into clinical indicators and combining the indicators into the Textbook Outcome approach, this data can also be used for monitoring quality of care.
For the BMJ Open publication, 45,000 patients from 62 Dutch hospitals were analyzed who had undergone gastrointestinal liver treatments. These concerned gallstone disease, inflammatory bowel disease and colorectal cancer screening. For each patient, it was individually determined if ‘Textbook Outcome’ was met with, based on five relevant and condition specific criteria. It then appeared to be possible to use this to measure variance between hospitals and to identify best practices, in order for providers to learn from each other.
Dr. Marc Verhagen, gastroenterologist and hepatologist at Diakonessenhuis Utrecht, explains: “We are trained for delivering the best care on a daily basis. Some interventions are covered by clinical registries for monitoring the quality of care. This new approach enables us to expand quality measurements to other interventions in short notice.” Dr. Rolf Bremmer from LOGEX adds: “This study for the first time shows that data from the basic administration in hospitals can profitably be used for providing automatized insight into quality of care. This is truly an international milestone, and possible because Dutch healthcare and its standards for administration are of a high level. This new model will also prove suitable for surgical, urological and cardiological interventions. Furthermore, this model will improve insights into quality of care and as such will be an important step towards value-based benchmarking and value-based contracting in the Netherlands and beyond.