From the Journals

Administrative data can help drive quality improvement in breast cancer care


 

FROM THE JOURNAL OF ONCOLOGY PRACTICE

Quality indicators of breast cancer care were successfully computed for more than 15,000 incident cases of breast cancer using electronic administrative databases in a project led by five regional oncology networks in Italy.

The project has shown that, despite some limitations in the use of administrative data to measure health care performance, “evaluating the quality of breast cancer care at a population level is possible,” investigators reported in the Journal of Oncology Practice.

The data obtained “from multiple administrative databases gathered in a real-world setting across five Italian regions” highlighted regional variations in breast cancer care and ways in which clinical guidelines were being overlooked, they wrote.

In doing so, the project confirmed that administrative data is “suitable” for measuring performance in health care and potentially useful for guiding quality improvement interventions. For instance, the project identified extensive use of blood tumor markers in breast cancer follow-up, wrote Valentina Guarneri, PhD, MD, of the University of Padova (Italy) and the Istituto Oncologico Veneto, also in Padova, and coauthors.

Oncologists and epidemiologists from the Italian regional oncology networks identified 46 clinically relevant indicators (9 structure indicators, 29 dealing with process, and 8 outcome indicators) by comparing pathways of care established by each network and identifying commonalities.

Of the 46 indicators, 22 were considered by the project leaders to be “potentially computable” from information retrieved by regional administrative databases. And of these 22 designed to be extractable, 9 (2 indicators of structure and 7 of process) were found to be actually evaluable for 15,342 cases of newly diagnosed invasive and/or in situ breast cancer diagnosed during 2016.

Blood tumor markers were tested in 44.2%-64.5% of patients in the first year after surgery – higher than the benchmark of 20% or less that was established to account for stage IV patients and other specific conditions in which markers might be indicated. National, international, and regional guidelines “discourage the use of blood tumor markers” in breast cancer follow-up, the investigators wrote.

The extensive use of these markers – observed across all five regions – is “a starting point to understanding how to improve clinical practice,” they added.

Other quality indicators that were evaluable included radiotherapy within 12 weeks after surgery if adjuvant chemotherapy is not administered (42%-83.8% in the project, compared with the benchmark of 90% or greater) and mammography 6-18 months after surgery (administered in 55.1%-72.6%, compared with the benchmark of 90% or greater), as well as the proportion of patients starting adjuvant systemic treatment (chemotherapy or endocrine therapy) within 60 days of surgery (for patients receiving systemic treatment).

To calculate the indicators, each regional cancer network used computerized sources of information including hospital discharge forms, outpatient records of diagnostic and therapeutic procedures, prescriptions of drugs reimbursed by the National Health Service in the hospital and outpatient settings, regional health registries, and the regional mortality registries.

All data used in the project came from regional repositories, which collect data from all National Health Service providers in the region, and not from single institutional repositories, the investigators noted.

More than half of the indicators expected to be assessable – but not found to be – were not computable as a result of data being unavailable (for example, pathology data) or incomplete, and as a result of data not being reliable for various reasons. The fact that examinations paid for directly by patients are not reported by the management systems of the National Health System was another complicating factor, they reported.

The authors disclosed funding and relationships with various pharmaceutical companies. The research was supported by the Periplo Association.

SOURCE: Guarneri V et al. J Oncol Pract. 2019 Dec 19. doi: 10.1200/JOP.19.00466.

Next Article: