(Wednesday, 8/1/2018) 10:30 AM - 11:00 AM
Room: Exhibit Hall | Forum 4
Purpose: Cancer patients undergoing radiotherapy (RT) may result in emergency department visits (EDV) due to the progression of the disease or treatment related toxicities. Predicting risks of EDV may lead to better clinical management and decreasing overall cost. The purpose of this work is to use Big Data tools to characterize risks of EDV on various cancer sites receiving RT only.
Methods: From 1/1/2012-12/31/2013 and 1/1/2015-6/30/2017 at a high volume center (HV) and three medium volume centers combined (MV) in our enterprise, EMR data were queried, including EDVs from the initiation to 60 days after RT with exclusion of concurrent chemotherapy. The EDV rates were stratified by RT volumes and disease sites. Fisher exact test was used for statistical analysis.
Results: A total of 5310 RT alone courses were completed. Of these patients, 95 (1.79%) EDVs were recorded. For the HV and MV centers, the RT alone courses were 2864 and 2446, with 25/0.87% and 70/2.9% EDVs, respectively. The HV center has lower EDV rate (p<0.01). Anal/Rectal (AR) and Gynecological (GYN) cancers had the highest overall EDV rates of 8.8% and 10.0%. For HV and MV centers, the number of RT courses and EDV rates for AR were 98/3.1% and 107/14%, respectively. The difference was significant (p<0.01). The number of RT courses and EDV rates for GYN were 38/0% and 52/11.5%, respectively. The difference was not statistically significant (p=0.08).
Conclusion: The ER visits rate for RT only patients is low, but high volume centers tends to have lower rate than the low volume centers. Using Big Data tools, we can monitor ER visits for patients receiving RT only, further predicting risking factors and preventing unexpected ER visit while providing proactive care for high risk patients.
Keywords
Not Applicable / None Entered.
Taxonomy
TH- Dataset analysis/biomathematics: Informatics
Contact Email
