Abstract
Rationale and objectives
The COVID-19 pandemic led to the national shutdown and subsequent reopening of cancer screening programs. Our diverse inner-city lung cancer screening program serves patients in the Bronx NY, which was severely affected by COVID-19, with the highest mortality in New York State in the spring of 2020. Staffing redeployment, quarantine protocols, increased safety measures, and changes in follow up resulted. The purpose of this study is to analyze the effect of the pandemic on lung cancer screening volumes during the first year of the pandemic.
Methods and materials
Retrospective cohort comprised of all patients enrolled in our Bronx, NY lung cancer screening program from March 2019 to March 2021 who underwent LDCT or appropriate follow-up imaging. The pre-pandemic and pandemic period were defined as 3/28/2019 to 3/21/2020 and 3/22/2020 to 3/17/2021, respectively, dichotomized by the New York State lockdown.
Results
1218 exams were performed in the pre-pandemic period and 857 in the pandemic period, a 29.6% decrease. The percentage of exams performed on newly enrolled patients decreased from 32.7% to 13.8% (p < 0.001). Patients in the pre-pandemic period and pandemic period respectively had the following demographic breakdown: mean age 66.9 ± 5.9 vs 66.5 ± 6.0, women 51.9% vs 51.6%, White 20.7% vs 20.3%, Hispanic/Latino 42.0% vs 36.3%. There was no significant difference in Lung-RADS scores for pre-pandemic and pandemic exams (p > 0.05). In the pandemic period, exam volume followed an inverted parabolic pattern, reflecting Covid surges for the cohort and all demographic subgroups.
Conclusion
The COVID-19 pandemic significantly decreased lung cancer screening volume and new enrollment in our urban inner-city program. Screening volumes demonstrated a parabolic curve reflecting pandemic surges following the initial wave, unlike other reports. The combination of the impact of COVID on our population and lack of staffing redundancy in the screening program, in the face of typical COVID isolation and quarantine absences, impeded early pandemic rebound of our lung cancer screening program. This highlights the necessity of fostering resilience by developing robust programmatic resources.
Keywords: COVID-19, Lung cancer, Cancer screening
1. Introduction
Lung cancer is the second most common cancer and the leading cause of cancer death in the United States. Lung cancer screening has been proven to substantially reduce lung cancer mortality. The Coronavirus Disease of 2019 (COVID-19) pandemic significantly impacted all facets of healthcare, including the national shutdown and reopening of various cancer screening programs. In a recent study, Chen et al. estimated that the COVID-19 pandemic caused a deficit in approximately 9.3 million cancer screening studies from March to July 2020, and only included breast, colorectal and prostate cancer screening.1 The lung cancer screening program at our urban academic medical center in the Bronx, NY was particularly impacted.
The population of patients in the Bronx faces unique challenging burdens. The Bronx had a household poverty rate of 26.4% in 2019, significantly higher than in the other four boroughs and 10 percentage points higher than the citywide average.2 According to a 2018 report by the Office of the New York State Comptroller, 60% of Bronx households devoted at least 30% of their incomes to rent (the level at which rent is considered a burden), higher than in any other borough and six percentage points higher than the citywide share.3
Our lung cancer screening program in the Bronx was additionally affected by staffing redeployment, quarantine protocols, increased safety and social distancing measures, and modifications in patient follow up. As the pandemic continued to evolve throughout 2020 and into 2021, our screening program continued to adapt.
Recent modeling studies have described the potential harm of delaying cancer diagnoses.4., 5., 6. A significant decline in cancer screening studies and biopsies is expected to reduce cancer diagnosis rates in the short term. However, in the long term, the healthcare community can expect increased cancer diagnosis rates, a larger proportion of later stage disease, greater cancer-related mortality rates and more years of life lost.5 Recent reports have described the effect of the pandemic on lung cancer screening programs.7., 8., 9.
Our lung cancer screening program serves a population that was severely impacted by COVID-19. The Bronx had the highest COVID mortality rate of all New York City and New York State counties in the spring of 2020, during the first wave of COVID-19.10 The purpose of the present study is to quantify the effects of the COVID-19 pandemic on lung cancer screening volumes during the first year of the pandemic in our understudied diverse, urban population. While we expected to see a sharp decrease in screening during the first wave of the pandemic, we hypothesized that there would be a persistent decrease in screening volume during the first pandemic year, without a full return to baseline.
2. Materials and methods
A retrospective analysis was conducted on the prospectively maintained database of our lung cancer screening program, accredited by the American College of Radiology (ACR). This study was approved by the institutional review board. The requirement to obtain informed consent was waived.
Our lung cancer screening program is part of a large urban academic medical center in a majority-minority inner city setting of low socioeconomic status in the Bronx, NY. The program was initiated in December 2012. Patient recruitment occurs through local physician referral. Eligibility using CMS (Centers for Medicare and Medicaid Services) criteria was verified by the program's full-time, bilingual (English and Spanish) nurse practitioner coordinator and part time clerk. CMS eligibility criteria were required for screening. The nurse practitioner's role was to conduct decision-making visits for lung cancer screening, review all CT results, communicate with referring providers, provide patient navigation for Lung-RADS 3 and Lung-RADS 4, offer smoking cessation referrals as needed, and to present at Thoracic Oncology Tumor Board. The part-time clerk/coordinator's role was to confirm patient eligibility, schedule CT scans, ensure insurance authorization for the CTs, and database management. Data, including demographic characteristics, medical comorbidities, imaging findings, interventions, and lung cancer diagnoses, were collected prospectively.
The present study cohort consists of patients enrolled in our lung cancer screening program who underwent LDCT (low dose computed tomography) for lung cancer screening and patients who underwent chest CT for screening, follow-up, or clinical indications at our institution from November 2018 to March 2021.
The date of lockdown in New York State (March 22, 2020) was used to dichotomize the timeline into pre-pandemic (March 28, 2019 to March 21, 2020) and pandemic (March 22, 2020 to March 17, 2021) periods. Monthly mean CT volumes were compared between these periods. LDCTs were exclusively interpreted by members of the Cardiothoracic Radiology division, a group of fellowship-trained, board-certified chest radiologists who reported the results according to the ACR Lung-RADS framework. During the study period, this group comprised 5 cardiothoracic fellowship-trained, board-certified, radiologists ranging from 2 to >30 years experience. Lung cancer diagnosis and mortality data were acquired from direct contact with patients' families, referring physicians, and chart review.
Data distribution was characterized by means and standard deviations for continuous data and frequencies for categorical data. Fisher exact test was used to determine statistical significance between categorical data. Student's t-test was utilized to calculate statistical significance for continuous data. Statistical analyses were performed using Microsoft Excel 2019 (Microsoft Corporation, Redmond, Washington).
3. Results
The screening cohorts in both the pre-pandemic and pandemic periods were diverse and gender-balanced (Table 1 ). A total of 1218 exams were performed in the pre-pandemic period and 857 exams in the pandemic period, representing a 29.6% decrease. There was no statistically significant difference in any demographic characteristic between exams in the pre-pandemic and pandemic periods. There were 400 newly enrolled patents during the pre-pandemic period and 118 in the pandemic period. The percentage of newly enrolled patients within each period significantly decreased from 37.4% to 16.5% (Fig. 1 ; p < 0.001).
Table 1.
Baseline demographics of pre-pandemic and pandemic lung cancer screening patients.
| Characteristic | Pre-pandemic | Pandemic | p-Value |
|---|---|---|---|
| Cases | 1218 | 857 | – |
| Age, mean ± SD | 66.9 ± 5.9 | 66.5 ± 6.0 | 0.12 |
| Male, n (%) | 586 (48.1%) | 415 (48.4%) | 0.93 |
| Race, n (%) | |||
| White | 252 (20.7%) | 199 (23.2%) | 0.38 |
| Black | 340 (27.9%) | 269 (31.4%) | 0.27 |
| Asian | 22 (1.8%) | 13 (1.5%) | 0.75 |
| Other | 476 (39.1%) | 289 (33.7%) | 0.2 |
| Unknown | 128 (10.5%) | 87 (10.2%) | 0.87 |
| Ethnicity, n (%) | |||
| Hispanic/Latino | 511 (42.0%) | 311 (36.3%) | 0.19 |
| Non-Hispanic/Latino | 547 (44.9%) | 439 (51.2%) | 0.19 |
| Unknown | 160 (13.1%) | 107 (12.5%) | 0.79 |
| Pack years smoking, mean ± SD | 48.6 ± 58.5 | 49.3 ± 68.0 | 0.84 |
Fig. 1.
Comparison of the proportion of new patient enrollment in the pre-pandemic vs pandemic period.
In the pre-pandemic period, an average of 101.5 exams occurred per month. Monthly volume decreased to an average of 71.4 exams per month in the pandemic period. Overall, case volume decreased to a minimum in the initial New York State lockdown in April and May 2020, with 14 and 10 exams per month performed, respectively. From June 2020 to November 2020, the exam volume returned to a pre-pandemic baseline, with a peak of 131 exams in September 2020. The exam volume subsequently declined corresponding to the second COVID-19 wave from December 2020 to March 2021, with a trough of 43 exams per month. This inverted parabolic pattern was true across all demographic groups in the pandemic period (Fig. 2 ), reflecting local COVID surges which impacted not only patients but staffing of the screening program. During the initial surge the NP coordinator was deployed to care for COVID patients, and the part time clerk was on family leave. This staff of two had multiple episodes of COVID isolation and quarantine throughout the remainder of the study period as required by local practice. There was no backup provided by the institution for the nurse practitioner or clerk during the pandemic. There was no option to work remotely.
Fig. 2.
Effect of COVID-19 pandemic on lung cancer screening program monthly exam volume: a) overall population pre- COVID-19 vs COVID-19 (3/28/2019–3/21/2020 vs. 3/22/2020–3/17/2021), b) men, c) women, d) white, e) non-white, f) Hispanic/Latino, g) non-Hispanic/Latino.
The distribution of Lung-Rads scores in the pre-pandemic period did not differ significantly from that of the pandemic period (Table 2 ; p = 0.91). More specifically, during the pre-pandemic period 7.6% were graded as Lung-Rads 4 compared to 7.9% during the pandemic period.
Table 2.
Lung-RADS score distribution.
| Score | Pre-pandemic |
Pandemic |
|||
|---|---|---|---|---|---|
| Count | Percentage | Count | Percentage | ||
| 1 | 677 | 55.6 | 462 | 53.9 | |
| 2 | 289 | 23.7 | 210 | 24.5 | |
| 3 | 160 | 13.1 | 117 | 13.7 | |
| 4 | 92 | 7.6 | 68 | 7.9 | |
| p = 0.91 | |||||
Seventeen patients (1.6% of total patients) were diagnosed with lung cancer in the pre-pandemic period compared to twelve (1.7%) in the pandemic period (Table 3 ; p = 0.92). 10/17 (58.8%) were diagnosed with Stage 1 cancer in the pre-pandemic period, while 4/12 were diagnosed with Stage 1 cancer in the pandemic period (33.3%). Conversely, 3/17 (17.6%) were diagnosed with Stage 4 cancer in the pre-pandemic period, while 3/12 (25.0%) were diagnosed with Stage 4 cancer during the pandemic period.
Table 3.
Clinical characteristics of lung cancer diagnoses.
| Characteristics | Pre-pandemic | Pandemic | p-Values |
|---|---|---|---|
| Number of patients diagnosed | 17 | 12 | 0.92 |
| Age at diagnosis (mean ± SD) | 66.9 ± 5.6 | 64.3 ± 7.6 | 0.31 |
| Male | 5 | 7 | 0.32 |
| Smoking status-current | 11 | 11 | 0.29 |
| Pack years (mean ± SD) | 46.4 ± 6.8 | 49.4 ± 13.3 | 0.49 |
| Method of diagnosis: Surgical biopsy | 14 | 3 | 0.05 |
| Method of diagnosis: Bronchoscopic biopsy | 2 | 5 | 0.23 |
| Method of diagnosis: Clinical | 1 | 0 | 0.58 |
| Method of diagnosis: Unknown | 0 | 4 | 0.1 |
| Time (days) from initial screening exam to diagnosis (mean ± SD) | 648.2 ± 650.4 | 591.2 ± 737.9 | 0.49 |
| Histology: Adenocarcinoma | 12 | 7 | 0.66 |
| Histology: Squamous cell carcinoma | 1 | 1 | 0.87 |
| Histology: Small cell | 2 | 3 | 0.55 |
| Histology: Unknown | 1 | 1 | 0.87 |
| Histology: Clinical dx | 1 | 0 | 0.58 |
| Stage: I | 10 | 4 | 0.38 |
| Stage: II | 1 | 3 | 0.34 |
| Stage: III | 2 | 1 | 0.85 |
| Stage: IV | 3 | 3 | 0.76 |
| Stage: Unknown | 1 | 1 | 0.87 |
| Disease status: No evidence of disease | 2 | 2 | 0.81 |
| Disease status: Stable | 4 | 1 | 0.49 |
| Disease status: Expired | 6 | 1 | 0.28 |
| Disease status: Advanced disease | 1 | 4 | 0.22 |
| Disease status: Active treatment | 0 | 4 | 0.1 |
| Disease status: Possible recurrence | 3 | 0 | 0.32 |
| Disease status: Nodules | 1 | 0 | 0.58 |
In both periods, the majority of patients diagnosed with cancer were current smokers, with a mean pack year history of 46.4 ± 6.8 and 49.4 ± 13.3 in the pre-pandemic vs pandemic period, respectively (p = 0.49). The most common method of cancer diagnosis in the pre-pandemic period was surgical biopsy, while in the pandemic period the most common method was bronchoscopic biopsy. Adenocarcinoma was the most common histological type in both periods. Cancers detected in the pre-pandemic period were predominantly stage I at the time of diagnosis. In contrast, the pandemic period had a relatively even proportion of both early and late-stage tumors.
4. Discussion
Lung cancer screening programs have been severely affected by the COVID-19 pandemic. Our program experienced a near-complete cessation of lung cancer screening during the first two months of the lockdown. Similar to our findings, Henderson et al. and Van Haren et al., describe a sharp reduction in number of screening chest CTs from March to June 2020.7., 9. Their subsequent recovery in volume of studies by July 2020 is comparable to most reports.1 However, these findings differ from our experience. While our institution did experience a rebound following the initial wave, it was not sustained through the end of 2020 and into 2021. There is very limited data published about how cancer screening programs, especially lung, evolved during the second half of 2020 and during 2021. Kriste et al. describe the volumes of breast cancer screening exams in an urban hospital in California, in which the number of mammograms similarly did not return to pre-pandemic levels.11
The difficulty in sustaining recovery during the second wave in our institution, is likely secondary to a combination of factors. Limited institutional resources with lack of redundancy in staffing of the screening program resulted in repeated short episodes of reduction and occasional suspension in scheduling patients for lung cancer screening in response to staff COVID quarantine and isolation requirements. Having only one full time nurse practitioner and one part-time clerk employed during the pandemic, without staffing redundancy, may have contributed just as much as other patient related factors. Similarly, staffing changes and redeployments within the thoracic surgery department impacted clinical practice and contributed to the preference of bronchoscopic biopsy, rather than surgical biopsy, for tissue diagnosis in the pandemic period.
The Bronx patient population incurs inherent socioeconomic and health disparities. In 2021, the Bronx was named the least healthy county in New York for the twelfth consecutive year.12 During the COVID pandemic, the Bronx population faced an exorbitant surge of unprecedented illness and death, with communities of low socioeconomical status disproportionally affected.13 All of these considerations may have played an important role when considering the impact of the COVID-19 pandemic on our lung cancer screening program. It is difficult to tease out whether staffing or patient-related factors had a larger impact.
Maintenance of programmatic quality metrics during the pandemic is supported by stable lung cancer diagnosis rates and Lung-RADS score distribution when comparing the pre-pandemic and pandemic period. However, we demonstrated a trend toward diagnosing early stage lung cancer less frequently during the pandemic period (Table 3). Van Haren et al. reviewed the impact of the initial wave of the pandemic on their lung cancer screening program, and found a significant increase in the percentage of Lung-RADS 4 results after screening resumed compared to prior (29% vs 8%, p < 0.01).7 Mynard et al. reported a significant increase in the proportion of patients with Stage 4 non-small cell lung cancer disease in the quarters following phased reopening.14 D'Ovidio et al., with respect to colorectal cancers, found that the “high-risk” adenomas detection rate was significantly higher in the “lockdown group” than in controls (47% vs. 25%) and colorectal cancers were more numerous in the “lockdown group” than in controls (8% vs. 1%) with the lockdown period (hazard ratio = 2.2) an independent predictor of high-risk adenomas and colorectal cancers.15 The “Stage Shift” phenomenon in cancer diagnoses secondary to the pandemic is a topic of ongoing study.16., 17., 18. A populational modeling study of >90,000 United Kingdom patients diagnosed with breast, colon, esophageal and lung cancer in the last decade estimates a striking increase in avoidable deaths secondary to the pandemic diagnostic delay, with up to 5.3% increase in deaths from lung cancer.4
This study is limited by its retrospective nature and access to documentation from a single multisite academic institution with inherent incomplete capture of relevant data. Additionally, our study lacks patient survey data that may provide additional context on what factors reduced volume. This data may be particularly helpful in further interpreting the reduction in volume during the second wave of the COVID-19 pandemic in the winter of 2020. The study was additionally limited by lack of follow-up beyond the defined pandemic period and we expect the impact of missed exams to be manifest over the long term. Finally, the moderate cohort size limited our statistical power to demonstrate differences between the time periods.
5. Conclusion
The COVID-19 pandemic significantly decreased lung cancer screening volume and new enrollment in our diverse urban inner-city program. Screening volumes demonstrated a parabolic curve reflecting pandemic surges following the initial wave, unlike other reports. The combination of the impact of COVID on our population and lack of staffing redundancy in the screening program, in the face of typical COVID isolation and quarantine absences, impeded early pandemic rebound of our lung cancer screening program. This highlights the necessity of fostering resilience by developing robust programmatic resources.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References
- 1.Chen R.C., Haynes K., Du S., Barron J., Katz A.J. Association of Cancer Screening Deficit in the United States with the COVID-19 pandemic. JAMA Oncol. 2021;7(6):878–884. doi: 10.1001/jamaoncol.2021.088. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Neighborhood profiles: the Bronx. Furman Center for Real Estate and Urban Policy. 2019. https://furmancenter.org/neighborhoods/view/the-bronx Accessed 6/21/2022.
- 3.DiNapoli T.P.B.K. An economic snapshot of the Bronx. 2018. https://www.osc.state.ny.us/files/reports/osdc/pdf/report-4-2019.pdf
- 4.Maringe C., Spicer J., Morris M., et al. The impact of the COVID-19 pandemic on cancer deaths due to delays in diagnosis in England, UK: a national, population-based, modelling study. Lancet Oncol. 2020;21(8):1023–1034. doi: 10.1016/S1470-2045(20)30388-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Alkatout I., Biebl M., Momenimovahed Z. Has COVID-19 affected cancer screening programs? A systematic review. Front Oncol. 2021;11 doi: 10.3389/fonc.2021.675038. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Yong J.H., Mainprize J.G., Yaffe M.J., et al. The impact of episodic screening interruption: COVID-19 and population-based cancer screening in Canada. J Med Screen. 2021;28(2):100–107. doi: 10.1177/0969141320974711. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Van Haren R.M., Delman A.M., Turner K.M., et al. Impact of the COVID-19 pandemic on lung cancer screening program and subsequent lung cancer. J Am Coll Surg. 2021;232(4):600–605. doi: 10.1016/j.jamcollsurg.2020.12.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Lang M., Yeung T., Shepard J.-A.O., et al. Operational challenges of a low-dose CT lung cancer screening program during the coronavirus disease 2019 pandemic. Chest. 2021;159(3):1288–1291. doi: 10.1016/j.chest.2020.10.045. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Henderson L.M., Benefield T., Bosemani T., Long J.M., Rivera M.P. Impact of the COVID-19 pandemic on volumes and disparities in lung cancer screening. Chest. 2021;160(1):379–382. doi: 10.1016/j.chest.2020.12.033. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Coronavirus COVID-19 death rates New York by county. Statista. 2021. https://www.statista.com/statistics/1109417/coronavirus-covid19-death-rates-new-york-by-county/
- 11.Krist A.H., Davidson K.W., Mangione C.M., et al. Screening for lung cancer: US preventive services task force recommendation statement. JAMA. 2021;325(10):962–970. doi: 10.1001/jama.2021.1117. [DOI] [PubMed] [Google Scholar]
- 12.County Health Rankings &. Roadmaps University of Wisconsin Population Health Institute. 2021. www.countyhealthrankings.org
- 13.Thompson C.N., Baumgartner J., Pichardo C., et al. COVID-19 outbreak - New York City, february 29-june 1, 2020. MMWR Morb Mortal Wkly Rep. 2020;69(46):1725–1729. doi: 10.15585/mmwr.mm6946a2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Mynard N., Saxena A., Mavracick A., Port J., Lee B., Harrison S., Chow O., Villena-Vargas J., Scheff R., Giaccone G., Altorki N. Lung cancer stage shift as a result of COVID-19 lockdowns in New York City, a brief report. Clin Lung Cancer. 2022 May;23(3):e238–e242. doi: 10.1016/j.cllc.2021.08.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.D'Ovidio V., Lucidi C., Bruno G., Lisi D., Miglioresi L., Bazuro M.E. Impact of COVID-19 pandemic on colorectal cancer screening program. Clin Colorectal Cancer. 2021;20(1):e5–e11. doi: 10.1016/j.clcc.2020.07.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Fedorenko C.R., Kreizenbeck K.L., Li L., Panattoni L.E., Shankaran V., Ramsey S.D. Stage at cancer diagnosis during the COVID-19 pandemic in western Washington state. J Clin Oncol. 2021;39(28_suppl) doi: 10.1200/jco.2020.39.28_suppl.145. 145-145. [DOI] [Google Scholar]
- 17.Flores R., Alpert N., Taioli E., McCardle K. Shift in lung cancer stage at diagnosis during the COVID-19 pandemic in New York City. Trans Lung Cancer Res. 2022;11(7):1514–1516. doi: 10.21037/tlcr-22-191. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Guven D.C., Sahin T.K., Yildirim H.C. Newly diagnosed cancer and the COVID-19 pandemic: tumor stage migration and higher early mortality [published online ahead of print, 2021 Oct 28] BMJ Support Palliat Care. 2021 doi: 10.1136/bmjspcare-2021-003301. bmjspcare-2021-003301. [DOI] [PubMed] [Google Scholar]