Initial and ten-year treatment patterns among 11,000 breast cancer patients undergoing breast surgery—an analysis of German claims data

Background We aimed to explore the potential of German claims data for describing initial and long-term treatment patterns of breast cancer patients undergoing surgery. Methods Using the German Pharmacoepidemiological Research Database (GePaRD, ~ 20% of the German population) we included patients with invasive breast cancer diagnosed in 2008 undergoing breast surgery and followed them until 2017. We described initial and long-term treatment patterns and deaths. Analyses were stratified by stage (as far as available in claims data), age at diagnosis, and mode of detection (screen-detected vs. interval vs. unscreened cases). Results The cohort comprised 10,802 patients. The proportion with neoadjuvant therapy was highest in patients < 50 years (19% vs. ≤ 8% at older ages). The proportion initiating adjuvant chemotherapy within four months after diagnosis decreased with age (< 50 years: 63%, 50–69: 46%, 70–79: 27%, 80 + : 4%). Among women < 69 years, ~ 30% had two breast surgeries in year one (70–79: 21%, 80 + : 14%). Treatment intensity was lower for screen-detected compared to interval or unscreened cases, both in year one (e.g., proportion with mastectomy ~ 50% lower) and within 2–10 years after surgery (proportions with radiotherapy or chemotherapy about one third lower each). Conclusions This study illustrates the potential of routine data to describe breast cancer treatment and provided important insights into differences in initial and long-term treatment by mode of detection and age. Supplementary Information The online version contains supplementary material available at 10.1186/s12885-022-09240-w.

: Visualization of the definition of mode of detection Table A1: Patient characteristics and initiation of first cancer treatment by stage at diagnosis

Measures to ensure a high data quality in the German Pharmacoepidemiological Research Database (GePaRD)
In GePaRD, we have established standard operating procedures and take several measures to ensure a high data quality. When we receive the data there are extensive plausibility checks including comparison with prior data years to detect inconsistencies or incompleteness in the data. Any issues are then resolved in dialogue with the respective health insurance provider. When we analyze the data, we take into account that there are types of data that have a high validity due to strict coding guidelines (hospital data) or automated processes for data entry (pharmacy data) but also types of data with a lower degree of validity such as outpatient diagnosis. When we use outpatient diagnoses, we therefore first develop algorithms, e.g., based on profile reviewing. These algorithms typically combine different types of information such as diagnosis codes and treatment data to increase specificity. In addition, we have conducted studies comparing incidences determined in GePaRD to those determined based on registries or other data sources, e.g.
with respect to information on tumor stage [1]. Finally, as far as possible regarding the strict regulations for data privacy in Germany, we try to conduct validation studies that directly link data from GePaRD to other data sources. For example, this has successfully been completed for information on death and date of death where information in GePaRD was linked to the official mortality data [2]. We classified a diagnosis as "screen-detected" if a screening mammography and a case conference were coded in the time period "two years before diagnosis until the quarter following the diagnosis". At least one of these codes had to be recorded in the surrounding quarters of diagnosis (i.e., in the quarter before diagnosis, in the quarter of diagnosis or the following quarter). The quarter following the diagnosis was also considered because codes may be recorded with a certain delay (this has been observed in case profiles, even though very rarely).
The diagnosis was classified as "interval cancer" if a screening mammography was coded in the time period "two years before diagnosis until the quarter following the diagnosis", and the criteria for "screen-detected" cancer were not fulfilled.
If no mammography screening was coded in the time period "two years before diagnosis until the quarter following the diagnosis" and the woman was eligible for screening (i.e. 50-69 years at diagnosis), this woman was classified as "unscreened but eligible". The remaining patients were classified as "unscreened and ineligible". Abbreviations: SD = standard deviation a Breast cancer was classified as "screen-detected" if a screening mammography and multidisciplinary case conference were coded in relevant time periods before and surrounding the diagnosis. It was classified as "interval-detected" if the woman had a screening mammography in the regular interval (two years) before diagnosis, but the criteria for "screen-detected" were not fulfilled. Patients without a screening mammography in the regular interval and aged 50-69 years at diagnosis were classified as unscreened, but eligible. The remaining patients were classified as "unscreened and ineligible". Some patients may be diagnosed, e.g., at age 70 and screened at age 69. b Within one year after diagnosis. Mastectomy includes those with both types of surgery. "Two or more surgeries" refers to additional breast conserving surgery/mastectomy in the first year after the first surgery. c This refers to adjuvant systemic therapy initiated within four months after breast surgery. d This refers to radiotherapy initiated within ten months after breast surgery.

Figure A2
Long-term treatment patterns (radiotherapy, cytostatic drugs, further surgery) and deaths in the years 2-10 among included breast cancer patients stratified by age group (