USE OF ELECTRONIC PATIENT DATA STORAGE FOR EVALUATION AND SETTING RISK CATEGORY OF LATE EFFECTS IN CHILDHOOD CANCER SURVIVORS
Rajala, Samuli (2020-02-25)
Julkaisu on tekijänoikeussäännösten alainen. Teosta voi lukea ja tulostaa henkilökohtaista käyttöä varten. Käyttö kaupallisiin tarkoituksiin on kielletty.
Julkaisun pysyvä osoite on:
There are currently over 4 600 adult childhood cancer survivors in Finland. Though survival rates are high, it is known that only one third of the survivors stay as healthy as their age-mates. Some of the late effects can occur quickly after the end of cancer treatment, but some may occur decades later. One of the growing challenges is to provide appropriate care for childhood and adolescent cancer survivors after transition to primary health care. Studies also show that primary health care physicists are self-doubting on their abilities to take care of childhood cancer survivors. Lack of a systematic follow-up plan may lead to excessive use of healthcare services and, on the other hand, delayed intervention. In agreement between national representatives of pediatric and adult oncology groups in Finland, international guidelines for determining the risk for late effects have been adapted to serve as a nation-wide guideline for the health care authorities. Turku University Hospital was given the task of developing an automatized system for calculating the risk of late effects based on electronic patient records saved on the Hospital Data Lake. For this purpose, an electronic algorithm that uses some details from exposure-based health screening guidelines published by the Children’s Oncology Group (COG) was created. The COG guidelines are based on the type of cancer, details of treatment modalities and cumulative doses of certain chemotherapies. The results gathered by the algorithm were compared with those manually calculated by a clinician. As a result, we got significant concordance between the manual and algorithm-based risk classification. A total of 355 patients received a classification through the algorithm and 325 of those matched with the manual categorization producing a Cohen coefficient of 0.91 (95% CI 0.88-0.95). Our findings indicate that automated algorithms can be used to efficiently and reliably categorize cancer survivors into late effect risk groups. This further enables automatized compilation of appropriate individual late effect follow-up plan for each survivor.