Shift of radiotherapy use during the first wave of the COVID-19 pandemic? An analysis of German inpatient data

Abstract

Objective: To assess the change in inpatient radiotherapy related to COVID-19 lockdown measures during the first wave of the pandemic in 2020.

Methods: We included cases hospitalized between January 1 and August 31, 2018-2020, with a primary ICD-10 diagnosis of C00-C13, C32 (head and neck cancer, HNC) and C53 (cervical cancer, CC). Data collection was conducted within the Medical Informatics Initiative. Outcomes were fractions and admissions. Controlling for decreasing hospital admissions during holidays, calendar weeks of 2018/2019 were aligned to Easter 2020. A lockdown period (LP; 16/03/2020-02/08/2020) and a return-to-normal period (RNP; 04/05/2020-02/08/2020) were defined. The study sample comprised a control (admission 2018/19) and study cohort (admission 2020). We computed weekly incidence and IR ratios from generalized linear mixed models.

Results: We included 9365 (CC: 2040, HNC: 7325) inpatient hospital admissions from 14 German university hospitals. For CC, fractions decreased by 19.97% in 2020 compared to 2018/19 in the LP. In the RNP the reduction was 28.57% (p < 0.001 for both periods). LP fractions for HNC increased by 10.38% (RNP: 9.27%; p < 0.001 for both periods). Admissions for CC decreased in both periods (LP: 10.2%, RNP: 22.14%), whereas for HNC, admissions increased (LP: 2.25%, RNP: 1.96%) in 2020. Within LP, for CC, radiotherapy admissions without brachytherapy were reduced by 23.92%, whereas surgery-related admissions increased by 20.48%. For HNC, admissions with radiotherapy increased by 13.84%, while surgery-related admissions decreased by 11.28% in the same period.

Conclusion: Related to the COVID-19 lockdown in an inpatient setting, radiotherapy for HNC treatment became a more frequently applied modality, while admissions of CC cases decreased.

Keywords: Admissions; Covid-19 pandemic; Germany; Lockdown; Radiotherapy.

Fig. 1

Radiotherapeutic fractions for malignant neoplasm of cervix uteri. Line charts of the cumulative weekly performed radiotherapeutic fractions related to malignant neoplasms of cervix uteri across all 14 participating sites for the study cohort (blue) and the control cohort (yellow) from January 13 (adjusted week -9) to August 16 (adjusted week 21) 2020. a Overall radiotherapeutic fractions (OPS 8‑52*); b Megavoltage radiation therapy (OPS 8‑522*, 8‑523*); c Brachytherapy (OPS 8-524*, 8-525*). The shaded area represents the standard deviation of the weekly average of the years 2018 and 2019. Additionally, the data are shown without time resolution as boxplots in supplemental Figure S1

Fig. 2

Hospital admissions for malignant neoplasm of cervix uteri. Line chart of the cumulative weekly hospital admissions related to malignant neoplasms of cervix uteri across all 14 participating sites for the study cohort (blue) and the control cohort (yellow) from January 13 (adjusted week -9) to August 16 (adjusted week 21) 2020. The shaded area represents the standard deviation of the weekly average of the years 2018 and 2019. Additionally, the data are shown without time resolution as a boxplot in supplemental Figure S2

Fig. 3

Hospital admissions for malignant neoplasm of cervix uteri stratified by treatment groups. Line charts of the cumulative weekly hospital admissions related to malignant neoplasms of cervix uteri across all 14 participating sites for the study cohort (blue) and the control cohort (yellow) from January 13 (adjusted week -9) to August 16 (adjusted week 21) 2020 stratified by therapy categories. a Hospital admissions with radiotherapeutic procedures, without surgery-related procedures and without brachytherapy. b Hospital admissions with radiotherapeutic procedures, without surgery-related procedures and presence of brachytherapy. c Hospital admissions with radiotherapeutic procedures, without surgery-related procedures and presence of chemotherapy. d Hospital admissions with presence of surgery-related procedures. The shaded area represents the standard deviation of the weekly average of the years 2018 and 2019. Additionally, the data are shown without time resolution as boxplots in supplemental Figure S3

Fig. 4

Radiotherapeutic fractions for malignant neoplasm of head & neck. Line charts of the cumulative weekly performed radiotherapeutic fractions related to malignant neoplasms of head & neck across all 14 participating sites for the study cohort (blue) and the control cohort (yellow) from January 13 (adjusted week -9) to August 16 (adjusted week 21) 2020. a Overall radiotherapeutic fractions (OPS 8‑52*); b Megavoltage radiation therapy (OPS 8‑522*, 8‑523*). The shaded area represents the standard deviation of the weekly average of the years 2018 and 2019. Additionally, the data are shown without time resolution as boxplots in supplemental Figure S4

Fig. 5

Hospital admissions for malignant neoplasm of head & neck. Line chart of the cumulative weekly hospital admissions related to malignant neoplasms of head & neck across all 14 participating sites for the study cohort (blue) and the control cohort (yellow) from January 13 (adjusted week –9) to August 16 (adjusted week 21) 2020. The shaded area represents the standard deviation of the weekly average of the years 2018 and 2019. Additionally, the data are shown without time resolution as a boxplot in supplemental Figure S5

Fig. 6

Hospital admissions for malignant neoplasm of head & neck stratified by treatment groups. Line charts of the cumulative weekly hospital admissions related to malignant neoplasms of head & neck across all 14 participating sites for the study cohort (blue) and the control cohort (yellow) from January 13 (adjusted week –9) to August 16 (adjusted week 21) 2020 stratified by therapy categories. a Hospital admissions with radiotherapeutic procedures, without surgery-related procedures. b Hospital admissions with radiotherapeutic procedures, without surgery-related procedures and presence of chemotherapy. c Hospital admissions with presence of surgery-related procedures. The shaded area represents the standard deviation of the weekly average of the years 2018 and 2019. Additionally, the data are shown without time resolution as boxplots in supplemental Figure S6