The Impact of the COVID-19 Pandemic on Head and Neck Cancer Diagnosis in the Piedmont Region, Italy: Interrupted Time-Series Analysis

Abstract

Background: The COVID-19 pandemic has likely affected the most vulnerable groups of patients and those requiring time-critical access to healthcare services, such as patients with cancer. The aim of this study was to use time trend data to assess the impact of COVID-19 on timely diagnosis and treatment of head and neck cancer (HNC) in the Italian Piedmont region.

Methods: This study was based on two different data sources. First, regional hospital discharge register data were used to identify incident HNC in patients ≥18 years old during the period from January 1, 2015, to December 31, 2020. Interrupted time-series analysis was used to model the long-time trends in monthly incident HNC before COVID-19 while accounting for holiday-related seasonal fluctuations in the HNC admissions. Second, in a population of incident HNC patients eligible for recruitment in an ongoing clinical cohort study (HEADSpAcE) that started before the COVID-19 pandemic, we compared the distribution of early-stage and late-stage diagnoses between the pre-COVID-19 and the COVID-19 period.

Results: There were 4,811 incident HNC admissions in the 5-year period before the COVID-19 outbreak and 832 admissions in 2020, of which 689 occurred after the COVID-19 outbreak in Italy. An initial reduction of 28% in admissions during the first wave of the COVID-19 pandemic (RR 0.72, 95% CI 0.62-0.84) was largely addressed by the end of 2020 (RR 0.96, 95% CI 0.89-1.03) when considering the whole population, although there were some heterogeneities. The gap between observed and expected admissions was particularly evident and had not completely recovered by the end of the year in older (≥75 years) patients (RR: 0.88, 0.76-1.01), patients with a Romano-Charlson comorbidity index below 2 (RR 0.91, 95% CI: 0.84-1.00), and primary surgically treated patients (RR 0.88, 95% CI 0.80-0.97). In the subgroup of patients eligible for the ongoing active recruitment, we observed no evidence of a shift toward a more advanced stage at diagnosis in the periods following the first pandemic wave.

Conclusions: The COVID-19 pandemic has affected differentially the management of certain groups of incident HNC patients, with more pronounced impact on older patients, those treated primarily surgically, and those with less comorbidities. The missed and delayed diagnoses may translate into worser oncological outcomes in these patients.

Keywords: COVID-19; Italy; cancer; cancer detection; head and neck; head and neck cancers; interrupted time-series analysis (ITSA); time trend analyses.

Figure 1

Number of new SARS-CoV-2 infections (A), and COVID-19 hospital and intensive care unit (ICU) occupancy (B) per 1,00,000 resident population in Piedmont during 2020. A map of Italy with Piedmont location (C).

Figure 2

Temporal trends in incident HNC in the Piedmont Region, Italy. (A) Temporal trends in observed monthly HNC cases in Piedmont (red line) and the predicted number of cases estimated on the data from the pre-pandemic period (green solid and gray dashed line). (B) Cumulative observed and predicted HNC cases in 2020. Cumulative predicted HNC cases for 2020 are estimated on the data from the pre-pandemic period (2015–2019).

Figure 3

Temporal trends in incident HNC in the Piedmont Region, Italy, by age group. Temporal trends in observed monthly and cumulative HNC cases in Piedmont (red line) and the predicted number of cases estimated on the data from the pre-pandemic period with a linear term for time and Fourier terms to model holiday-related seasonality in the HNC detection (green line upper panel; blue line lower panel). Vertical red dashed line indicates the start of the COVID-19 outbreak in Italy with national restrictions.

Figure 4

Temporal trends in HNC diagnosed at late and early stage among patients eligible for the HEADSpAcE study recruitment. Disease stage at diagnosis (TNM Classification of Malignant Tumors, AJCC cancer staging manual, VIII ed): early-stage (I and II) and late-stage (III and IV).