key: cord-302758-i5pe61h1
authors: Pier, Matthew M.; Pasick, Luke J.; Benito, Daniel A.; Alnouri, Ghiath; Sataloff, Robert T.
title: Otolaryngology-related Google Search trends during the COVID-19 pandemic
date: 2020-06-19
journal: Am J Otolaryngol
DOI: 10.1016/j.amjoto.2020.102615
sha: 
doc_id: 302758
cord_uid: i5pe61h1

OBJECTIVE: To assess trends of Google Search queries for symptoms and complaints encountered commonly in otolaryngology practices during the coronavirus disease 2019 (COVID-19) pandemic when in-person care has been limited. MATERIALS AND METHODS: In this cross-sectional study, data on Google Search queries in the United States for 30 otolaryngology-related terms were obtained from Google Trends. The means of relative search volume from the COVID-19 period (March 29, 2020 through May 16, 2020) were compared to similar periods from 2016 to 2019 using a t-test of two independent samples. RESULTS: In total, 16.6% of search terms had significant increases in relative search volume during the COVID-19 period, with the largest percentage increase for “can't smell” (124.4%, p = .006), followed by “allergies” (30.3%, p = .03), “voice pain” (26.1%, p = .008), and “ears ringing” (19.0%, p < .001). Of all search terms, 26.7% had significant decreases in relative search volume, including the largest percentage decrease for “laryngitis” (59.8%, p < .001), followed by “thyroid nodule” (54.4%, p < .001), “thyroid cancer” (45.6%, p < .001), and “ENT” (34.9%, p < .001). CONCLUSION: This study demonstrates that Google search activity for many otolaryngology-related terms during the COVID-19 pandemic has increased or decreased significantly as compared to previous years. With reduced access to in-office otolaryngology care in the United States during the COVID-19 pandemic, these are important considerations for otolaryngology practices to meet the needs of patients who lack access to care.

The World Health Organization declared the coronavirus disease 2019 (COVID-19) a global pandemic on March 11, 2020 [1] . In the United States (U.S.), federal guidelines for "social distancing" followed shortly thereafter, and by March 30, 30 states had officially announced "stay-at-home" orders, and 12 additional states issued similar orders during the following week [ [2] , [3] ]. Outpatient medical visits, including those in otolaryngology practices, have been reduced drastically secondary to these measures in an effort to protect patients and providers, as well as preserve personal protective equipment [4] . Changes in regulations have prompted providers to offer more telehealth opportunities, and some otolaryngology practices have increased utilization of this method of care delivery [ [4] , [5] , [6] ]. For patients unable to access in-person or telehealth care, it is plausible that they might increasingly utilize internet search engines such as Google Search (Google Inc., Mountain View, California) to obtain information about symptoms, diagnoses, or treatments. During this unprecedented situation, it is important to understand the alternative ways in which patients are obtaining medical information in order to develop strategies within the otolaryngology community for meeting their needs.

Google Trends (GT) is a free and publicly available tool that provides information on geospatial and temporal patterns in search volumes for user-specified terms [7] . GT has contributed to the emerging field of "infodemiology", showing utility in assessing outbreaks of influenza-like illness or providing unique insights into human behavior [ [8] , [9] , [10] ]. This tool also has been utilized to examine relationships of otolaryngology-related search terms with environmental factors [11] , [12] . This study aims to assess trends within the U.S. for Google Search queries of symptoms and complaints encountered commonly in otolaryngology practices comparing the time of COVID-19 pandemic with similar time periods in previous years. 

GT determines the proportion of each search term among an anonymized sample of all search requests performed using Google Search [13] . Repeated searches by the same individual over a short period of time are eliminated. To account for geographical differences in total search volumes, GT normalizes each data point by dividing by the total searches within the time range and region. Relative Search Volume (RSV) is presented on a scale from 0 to 100 based on a term's proportion to all other searches, with the query's peak for the time range and geographic region set at 100 and all other data points divided by that peak value. For example, a value of 100 is the peak popularity for the term, while a value of 50 means that the term is half as popular. Table 1 includes the 30 search terms selected by the authors. These included a range of otolaryngology-related symptoms and complaints in technical and non-technical terminology as well as variations on a term to capture search activity more broadly. Internet search activity data were obtained and downloaded from the GT website (https://trends.google.com/). Each term search was performed with the following parameters: "United States" region; "1/1/2016 through 5/16/2020" time range; "All categories" for category; "Web search" type of search. Our aim was to investigate search volume coinciding with "stay-at-home" orders issued by the majority of states in the U.S. by March 30, 2020. 3 

RSV data from the periods in 2016-2019 were pooled, the means calculated, and the RSV means from the corresponding COVID-19 timeframe in 2020 were calculated. Comparison of the RSV means between these time periods was performed using a t-test of two independent samples assuming unequal variances with two-sided p-value<.05 considered statistically significant. All computations were conducted in Microsoft Excel 2017 (Microsoft Corporation, Redmond, Washington).

Comparisons of the means of RSV and percentage change for respective search terms are described in Table 2 . Of the 5 (16.6%) search terms with statistically significant increases in RSV during the COVID-19 period, the largest percentage increase was for "can't smell" Three pairs of non-technical and technical terms were examined: "ears ringing" and "tinnitus", "hoarseness" and "dysphonia", and "trouble swallowing" and "dysphagia". "Ears ringing" demonstrated a significant 19.0% increase in RSV (μ 1 p=.60), while "dysphagia" decreased significantly by 8.5% (μ 1 -μ 2 =-6.5 [95% CI -11.1 to -1.8];

p=.02).

During the COVID-19 pandemic when "stay-at-home" orders and "social distancing" guidelines drastically affected everyday life as well as otolaryngology practices in the U.S., we present evidence that internet search activity for many otolaryngology-related terms has changed significantly.

Of the 5 search terms with significant increases in RSV, "can't smell" demonstrated the greatest increase. A likely contributing factor to this increase was the inclusion of olfactory dysfunction as a possible symptom of COVID-19. By mid-March 2020, the American Academy of Otolaryngology-Head and Neck Surgery recommended including this symptom when screening for possible COVID-19 [14] . Meta-analysis later indicated a 52% prevalence of olfactory dysfunction among infected patients [15] . This was discussed widely in the media, J o u r n a l P r e -p r o o f 7 which likely contributed to the 124.4% increase in Google searches for "can't smell" seen in this study [16] . The four additional search terms with significant increases in RSV were "allergies", "voice pain, "ears ringing" and "ear pain". In attempting to understand this trend, it is interesting that all of these terms are non-technical and could be associated with patients being unable to attend appointments with otolaryngologists. As a quantitative example of appointment reductions in an otolaryngology practice, Kasle et al. reported a roughly 80% reduction in completed appointments from mid-March 2020 to mid-April 2020 compared with the corresponding period in 2019 [4] . Similarly, emergency departments have seen 80% reductions in the number of otolaryngology-related consultations during the COVID-19 pandemic [17] . It is possible that the increases in search queries for these four non-technical terms could be attributed partially to patients resorting to internet searches in lieu of medical consultation to obtain information about symptoms or treatment. However, this association is speculative and warrants further investigations such as correlating GT searches for "allergies" with regional pollen concentrations [18] . Nonetheless, this finding highlights the increasing use of the internet for certain nontechnical symptoms during the COVID-19 period. With the marked variability in the quality of internet information on treatments for common otolaryngologic problems, this highlights the need for improvements to the web-based information available to patients [19] .

Of the 30 terms investigated, 26.5% demonstrated significant decreases in RSV during the COVID-19 period. In assessing the possible factors contributing to these trends, it may be important to recognize the 34.4% reduction in RSV for "ENT" as the context for the decreases seen in the other terms. This decrease in "ENT" queries could reflect the reduction in referrals to otolaryngologists and the widespread cancellations of elective surgeries [20] . Previous studies have shown that patients utilize more internet search terms after receiving a diagnosis as J o u r n a l P r e -p r o o f 8 compared to before it has been given [21] . With fewer patients accessing otolaryngologists to receive diagnoses such as thyroid cancer or dysphonia, a decreased post-diagnosis search phenomenon could explain the reductions in certain search queries. This is further supported by the findings for the three pairs of non-technical and technical terms compared in this study. The only significant RSV increase for these was the non-technical term "ears ringing", while the only significant RSV decreases occurred with the technical terms "dysphonia" and "dysphagia." RSV reductions for these terms, as well as "thyroid nodule", could provide insight into patients' diagnostic needs that may have been addressed insufficiently during this period. Prompt attention to this matter may be warranted, as a portion of these diagnoses may represent underlying cancer and delays in care for head and neck cancer are associated with decreased survival [22] .

period to compensate for the reduction of in-office encounters, but this degree of technological adaptation has not been seen nationally [ [4] , [23] ]. There is a substantial amount of literature addressing the unique challenges to implementing telehealth in otolaryngology practices, and these GT findings could be incorporated into these strategies to meet patients' needs both during the COVID-19 pandemic and after it has resolved [ [24] , [25] , [26] ]. Looking beyond the ongoing pandemic, GT represents a potentially powerful source of insight into population-level searches for medical information that could guide timing of public health policies and improve understanding of regional differences in otolaryngology-related patient needs.

This report has limitations. The selected terms that were investigated represent a subjective bias toward terms that the authors suspected to be searched commonly on the internet.

GT does not provide quantitative information on search terms. Therefore, the observed changes in this study cannot be quantified to determine actual numbers of searches. Data are reported J o u r n a l P r e -p r o o f 9 from a population sample determined by GT, and thus may not represent accurately the entire population of the region under investigation. Additionally, there is inherent bias when using GT data as they are generated by a population that is literate, technologically competent, has access to the internet, and has selected Google as a search engine. The underlying motivations to perform a Google search vary widely and could include obtaining information on symptoms experienced by individuals, responding to media coverage, seeking information for academic interests, or many other motivations. These motivations cannot be discerned directly from GT data, which limits the interpretation of the information. Nevertheless, the trends suggest that patients are responding to COVID-19-related restrictions in otolaryngology care by changing their patterns of self-education through the internet.

The COVID-19 pandemic has challenged the ability of otolaryngologists to provide care to many patients in the U.S. This study demonstrates that Google search activity for many otolaryngology-related terms during this period has increased or decreased significantly as compared to previous years. These trends may suggest unmet needs of patients that otolaryngology practices should consider and attempt to address through expanded telehealth utilization. It also suggests that otolaryngologists may need to be prepared to address the information and misinformation that patients have acquired, which is likely to affect the substance of future in-person or telehealth encounters. Additional investigations are needed to better understand how patients utilize the internet to obtain medical information, to improve the quality of available web resources, and to determine the effects of such self-education on patients' future behaviors in caring for their otolaryngologic disorders.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Relative search volume "Can't smell" "Allergies"

Pandemic March: COVID-19's first wave circumnavigates the globe

Coronavirus Guidelines for America | The White House. The White House

When State Stay-at-Home Orders Due to Coronavirus Went into Effect

Outpatient Otolaryngology in the Era of COVID-19: A Data-Driven Analysis of Practice Patterns. Otolaryngol Head Neck Surg

Covid-19 and Health Care's Digital Revolution

Telehealth Opportunities for the Otolaryngologist: A Silver Lining During the COVID-19 Pandemic. Otolaryngol Head Neck Surg

Google Trends. Google Trends

Assessing the Methods, Tools, and Statistical Approaches in Google Trends Research: Systematic Review

Correlation between national influenza surveillance data and google trends in South Korea

Forecasting type-specific seasonal influenza after 26 weeks in the United States using influenza activities in other countries

Doctor Google: Correlating internet search trends for epistaxis with metropolitan climates

Effect of environmental factors on Internet searches related to sinusitis

FAQ about Google Trends data -Trends Help

Anosmia, Hyposmia, and Dysgeusia Symptoms of Coronavirus Disease. American Academy of Otolaryngology-Head and Neck Surgery

The Prevalence of Olfactory and Gustatory Dysfunction in COVID-19 Patients: A Systematic Review and Meta-analysis

Otolaryngol Head Neck Surg

Doctors warn an inability to smell could be a symptom of Covid-19

COVID-19: what happened to all of the otolaryngology emergencies?

Google Trends and pollen concentrations in allergy and airway diseases in France

How good is Google? The quality of otolaryngology information on the internet

Evolving Management of COVID-19: A Multiinstitutional Otolaryngology Perspective. Otolaryngol Head Neck Surg

Online information seeking behaviors of breast cancer patients before and after diagnosis: From website discovery to improving website information

Quantitative survival impact of composite treatment delays in head and neck cancer

COVID-19 transforms health care through telemedicine: evidence from the field

Smartphone-enabled wireless otoscope-assisted online telemedicine during the COVID-19 outbreak

Embracing telemedicine into your otolaryngology practice amid the COVID-19 crisis: An invited commentary

Diagnosis and Treatment of Voice Disorders

Weekly Google Search results for "Laryngitis" and "Thyroid nodule" a a time periods analyzed are in bold, which include the COVID-19 period

Relative search volume "Laryngitis