Detailed of the project
The Kaduceo team was asked by the team of the Center dedicated to rare respiratory diseases of the CHI of Créteil to carry out a comparison of care pathways between patients with PCD and patients with cystic fibrosis.
Primitive Ciliary Diskinesia (PCD) has long been considered a less serious disease than Cystic Fibrosis. However, patients with PCD present a significant morbidity and require a medical management similar to that of patients with Mucoviscidosis. The care pathway for patients with MCI is therefore often less structured and there are few referral centers in this field.
Since 2004, the CHIC has been home to a Center for Cystic Fibrosis, called “CRCM” (Center of Resource and Competence for Cystic Fibrosis). The doctors would like to draw inspiration from the management of Muco patients to enrich the management of PCD patients. Hence the need to compare the care pathways associated with these two pathologies.
We sought to identify the patterns common to both pathways as well as the flaws in the DCP pathway. In addition, CHIC physicians need indicators on these two care pathways to justify the creation of a center specialized in rare respiratory diseases that would allow for the joint management of both diseases.
Data set from the medico-administrative PMSI database of the CHI Créteil including all consultations and stays made by the patients. With the help of the Medical Information Department (DIM), we enriched these data with the consultations of specialists seen by patients during day hospitalizations.
Finally, these hospital data were completed by a clinical database listing the socio-demographic characteristics of the patients.
Solutions provided by Kaduceo
- Identification of the components of the care pathways for patients with Muco and PCD
- Highlighting of the elements common to both pathways and the differences that should be expected to be observed
- Differentiation of phenomena related to coding issues from those related to medical practices in place at CHIC
- Definition of different indicators to be able to quantitatively compare the two paths
- Development of new algorithms for calculating the frequency of appointments, taking into account the limitations of the PMSI
- Proposal of a cost estimate for each of the two pathways based on PMSI billing data
Result and further work
- Identification of points of improvement in the DCP care pathway (physiotherapist follow-up, HDJ)
- Detection of potential HDJ
- Scientific arguments to justify the interest of creating a center dedicated to rare diseases
Based on the data science work carried out by Kaduceo, the practitioners involved in the care of patients with PCD were able to optimize their management. The management of the CHI Créteil establishment was also able to rely on the results of the analyses conducted to organize the care activity.
A center dedicated to the management of patients with PCD has existed at the Créteil Hospital since 2007. The collaboration with Kaduceo and the comparative analysis of the pathways allowed the practitioners to justify the installation of this center in a single building, in order to facilitate cooperation between the different specialists involved in the care, to facilitate access to different specialists for PCD patients…
In the long term, this new organization will make it possible to increase the number of HDJ for DCP patients. This means fewer hospital visits for the patients and higher revenues for the institution.
Following the adjustments in the management of PCD patients, new studies should be conducted by the data scientists of the Kaduceo team in order to evaluate the impact on patient health. It will also be necessary to analyze the financial impact for the CHIC by organizing a DCP pathway along the same lines as the one for muco and thus favoring HDJ.
Other resources you may be interested in
No consensus exists about the doses of analgesics, sedatives, anesthetics, and paralytics used in critically ill neonates.
Kaduceo, co-author of work presented at the 24th International Workshop on Design, Optimization, Languages and Analytical Processing of Big Data