This article is more than 5 years old. Information may no longer be current.
Accurate sarcoma disease coding crucial to maintain valid databases
Click Here to Manage Email Alerts
Inaccurate coding may lead to underreporting of sarcoma cases in national databases, according to a research letter published in JAMA Oncology.
“At present, the net effect of coding errors is unknown,” Heather G. Lyu, MD, clinical fellow in surgery at Brigham and Women’s Hospital, and colleagues wrote. “Discussions among surgeons, pathologists, coders and tumor registrars about how to specify sarcomas are encouraged.”
Because sarcoma is rare, performing appropriately powered studies is challenging, and researchers often rely on tumor registries and population-based data to determine sarcoma incidence and outcomes. This increases the importance of accurate data collection, according to the researchers.
The analysis included 2,715 patients who underwent resection of primary or recurrent sarcoma between 2012 and 2016 by five sarcoma surgeons at Brigham and Women’s Hospital and Dana-Farber Cancer Institute.
Researchers confirmed diagnoses by comparing Brigham and Women’s Hospital operative diagnosis codes — which use ICD-9 and ICD-10 — with pathology reports, and classified them as true positive, false negative or true negative. Then, researchers used the true positive cohort to evaluate each patient’s Dana-Farber Cancer Institute diagnosis code — using ICD-O-3 — to determine the accuracy of their tumor registry.
Overall, 1,237 patients had a sarcoma diagnosis confirmed by pathologic findings.
Using the ICD-9 and ICD-10 codes, 764 those patients (61.8%) had cases that were accurately coded as sarcoma, 208 (16.8%) had a nononcologic diagnosis, and 265 (21.4%) had an organ site-based malignancy code.
Further, 487 of 855 patients (57%) treated by surgical oncologists and 277 of 382 patients (72.5%) treated by orthopedic oncologists had cases that were accurately coded.
The researchers found that organ-confined sarcoma was commonly coded with a nonsarcoma, organ-site code instead. For example, 31.4% of gastric gastrointestinal stromal tumors were coded as gastric cancer and 52.2% of breast angiosarcomas were coded as breast cancer.
Based on the ICD-O-3 codes, 631 of 1,055 patients (59.8%) were accurately coded, 26 (2.5%) were coded with another cancer diagnosis, and 398 (37.7%) were not listed in the registry.
Limitations of the study included the possibility that “coding inaccuracies as identified in this study may be specific to our institution,” the researchers wrote, adding the findings may not be generalizable for all sarcoma centers.
Lyu and colleagues noted, however, that tumor registrars receive uniform training by American Joint Committee on Cancer guidelines, meaning inaccurate coding could be widespread.
“This study emphasizes that the vague nature of definitions for diseases can lead to coding inaccuracies that can be propagated through datasets, which is an issue that possibly extends beyond any single institution,” Lyu and colleagues wrote. – by Trudi Gilfillian
Disclosures: The authors report no relevant financial disclosures.
Click Here to Manage Email Alerts