Introduction
COVID-19 has changed the landscape of cancer care, resulting in forgone screening, delayed diagnosis, and postponed treatment [1]. Healthcare professions were quick to adopt measures such as expedited treatment regimens, telemedicine, and PPE adherence to ensure high-quality cancer treatment [2]. However, there has been no study that has examined the interest in cancer care from patients’ perspectives. The objective of this study is to use Google Trends, a publicly-available search engine analytics tool, to assess how the public’s interest in cancer care, specifically radiotherapy, has evolved during the pandemic.
Materials and methods
Google Trends (Google LLC, Mountain View, CA) is a search engine analytics tool that analyzes the popularity of search queries, presented as “search volume index” (SVI): the search frequency of the queried term normalized to the highest frequency over specified time period and geographical location (scale of 0–100). The population included in Google Trends is a sample of all Google searches which is sufficient to be representative of the whole Google searches [3]. We queried Google Trends using the terms “IMRT”, “proton therapy”, “brachytherapy”, “SBRT”, “gamma knife”, “chemotherapy”, “cancer surgery”, and “radiotherapy”, within the United States during the period 1/5/2020–10/10/2020. Each term and its synonyms were queried and the one with the highest search volume was selected to represent said category. “Top search terms” and “Rising search terms” for each term were verified to confirm that there were no other confounding meanings.
Three time periods were established: pre-lockdown (1/5/20–3/14/20), lockdown (3/15/20–5/30/20), and reopening (5/31/20–10/10/20), representative of an ‘average/median’ lockdown time of each state [4]. Descriptive statistics (mean ± SD) were used to describe SVI. One-way ANOVA was used to compare mean values of the three periods. Linear regression (regression coefficient β) was used to examine SVI changes over time. Statistical analyses were performed using SPSS 25 (IBM Corp., Armonk, NY); significance was considered at p < 0.05.
Results
Comparing pre-lockdown with lockdown periods, statistically significant decreases in search volume were observed in all modalities (Fig. 1, Tab. 1). Comparing lockdown with reopening, statistically significant increases were observed except for stereotactic body radiation therapy (SBRT) and chemotherapy; these modalities demonstrated modest increases that were not statistically significant. Comparing pre-lockdown with reopening, statistically significant decreases were seen in all except intensity-modulated radiation therapy (IMRT) (65.45 vs. 56.74, p = 0.204), brachytherapy (70.10 vs. 66.05, p = 0.400), and gamma knife (69.10 vs. 57.53, p = 0.106).
|
|
Pre-lockdown |
Pre-lockdown v.s. lockdown |
Lockdown |
Lockdown v.s. reopening |
Reopening |
Pre-lockdown v.s. reopening |
|
Mean (SDa) |
p valueb |
Mean (SD) |
p value |
Mean (SD) |
p value |
|
|
IMRT |
65.40 (23.109) |
0.005 |
42.82 (12.734) |
0.039 |
56.74 (15.726) |
0.20 |
|
Proton therapy |
78.70 (12.157) |
< 0.001 |
55.45 (8.490) |
0.005 |
68.21 (12.255) |
0.02 |
|
Brachytherapy |
70.10 (16.835) |
0.005 |
54.09 (9.115) |
0.013 |
66.05 (10.768) |
0.40 |
|
SBRT |
66.10 (16.835) |
0.006 |
47.09 (9.628) |
0.256 |
54.95 (16.201) |
0.04 |
|
Gamma knife |
69.10 (15.308) |
< 0.001 |
39.36 (12.909) |
0.004 |
57.53 (17.138) |
0.11 |
|
Chemotherapy |
90.60 (5.502) |
< 0.001 |
76.82 (5.619) |
0.449 |
78.42 (5.480) |
< 0.001 |
|
Cancer Surgery |
90.40 (7.619) |
< 0.001 |
60.55 (7.607) |
0.000 |
74.37 (7.755) |
< 0.001 |
|
Radiotherapy |
86.70 (9.581) |
< 0.001 |
66.55 (6.502) |
0.033 |
72.53 (5.910) |
< 0.001 |
Linear regression analysis (Tab. 2) showed that brachytherapy (β = –1.213, p = 0.110) and chemotherapy (β = –1.130, p = 0.000) were least impacted by lockdown. During reopening, interest in proton therapy had the fastest and most robust return (β = 0.866, p = 0.001), followed by gamma knife, brachytherapy, IMRT and SBRT. Among the three branches of cancer treatment, surgery had the most robust return (β = 0.967, p = 0.000), followed by radiotherapy (β = 0.554, p = 0.020) and chemotherapy (β = 0.311, p = 0.183).
|
|
From pre-lockdown to nadira |
From nadir to after reopening |
||||
|
|
β |
R square |
p-value |
β |
R square |
p-value |
|
IMRT |
–5.406 |
0.527 |
0.008 |
0.636 |
0.118 |
0.07 |
|
Proton therapy |
–1.564 |
0.386 |
0.003 |
0.866 |
0.335 |
0.001 |
|
Brachytherapy |
–1.213 |
0.152 |
0.11 |
0.724 |
0.272 |
0.004 |
|
SBRT |
-1.274 |
0.270 |
0.01 |
0.483 |
0.087 |
0.11 |
|
Gamma knife |
-2.877 |
0.442 |
0.004 |
0.856 |
0.225 |
0.008 |
|
Chemotherapy |
–1.130 |
0.604 |
< 0.001 |
0.311 |
0.102 |
0.18 |
|
Cancer surgery |
–2.973 |
0.573 |
0.003 |
0.967 |
0.576 |
< 0.001 |
|
Radiotherapy |
–1.536 |
0.549 |
< 0.001 |
0.554 |
0.279 |
0.02 |
Discussion
Our findings suggest that public interest in cancer treatments decreased during lockdown and returned after reopening but, in general, is still lower than pre-lockdown levels. We postulate that public interest and response have largely been shaped by guidelines established by organizations.
SBRT has had statistically insignificant increases from lockdown low points after reopening. Slowed return in interest may reflect reduced number of diagnoses and delayed cancer treatment. Since SBRT plays an important role in treating early-stage non-small cell lung cancer [5], we postulate that the lower interest in SBRT may indicate decreased lung cancer screening and detection of early-stage disease, which is supported by what Kaufman et al. discovered using Quest diagnostics data [6].
Interest in brachytherapy had a change between pre-lockdown and reopening, was among the least affected by the lockdown, and showed a quick return to baseline after reopening. American Brachytherapy Society recommended minimizing delays for gynecologic brachytherapy [7]. Guidelines for brachytherapy during COVID-19 were also published, emphasizing the importance of timely brachytherapy for patients with gynecologic, breast, and prostate malignancies [8]. The observed trend was also supported by a retrospective cohort study by an individual U.S. institution, noting only minor treatment delays and interruptions for patients needing brachytherapy [9].
Gamma knife had a change when comparing pre-lockdown with reopening and the second-fastest rebound after reopening. Stereotactic surgeries (SRS) are frequently used for treating intracranial malignancies, benign tumor, and vascular malformations. Many of these can be treated electively, except primary or secondary intracranial malignancies and symptomatic lesions which should be treated on a semi-urgent basis [10]. Given its non-invasiveness, shorter course of treatment and no need for intensive care, SRS has been purported by many as an alternative for open surgery during the COVID-19 pandemic [10–12].
The fastest return after reopening was observed in proton therapy. This is in alignment with the goals stated in the guidelines published by multiple proton centers. The New York Proton Center and the MD Anderson Proton Therapy Center have respectively published their institutional COVID-19 guidelines regarding patient prioritization [13, 14]. Many of the conditions being treated (e.g., head and neck cancers, pediatric malignancies) cannot afford treatment delays. Moreover, many of the patients on proton therapy are on clinical trials; any delays in treatment risk violating the protocols.
Our study is limited by the short time-frame; longer trends may be more apparent with longer time elapsed. Future research could correlate this data with the public’s behavior when cancer screening/diagnosis/treatment data becomes available.
Conclusion
While it is promising to see an upward trend of interest in cancer treatment from lockdown levels, healthcare professionals should strive to provide timely cancer care, assuage patients’ fears of healthcare settings, and encourage patients to continue proper cancer screenings.
Conflict of interest
None declared.
Funding
None declared.
Author responsible for statistical analysis
Kaidi Wang, M.D. (e-mail: kwang@uams.edu).
Data sharing
Research data are available at https://trends.google.com/trends/?geo=US
Acknowledgements
This project was supported, in part, by the Translational Research Institute (TRI) at UAMS, grant UL1TR003107 through the National Center for Advancing Translational Sciences of the National Institutes of Health (NIH). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.