A retrospective study of a publicly available query tool that aggregates data on Google search trends was conducted. Google Trends is a useful tool for tracking the frequency of search terms over time. It can be used to analyze changes in public interest or awareness of certain topics, including breast cancer screening. With these data, researchers and health professionals gain insight into how people are searching for information about breast cancer screening and where the gaps in their knowledge may be.

Using Google Trends, different search terms related to breast cancer screening may be compared. This allows users to understand the search term that has garnered more interest from the public over a period of time. In addition, comparisons can also be made between different geographic regions, time frames, and categories.

The data for any search term are normalized to the time and location of a query by the division of the total searches of the geography and time range it represented, to compare relative popularity. The relative popularity for any term is reported as a relative search frequency (RSF) from 0 to 100, with 100 representing the peak popularity of a term.

A search for a single term gathers results that include the specific word queried. Next, a search of multiple terms includes each word in any order. A search for a term in quotes obtains results that include the specific order of words queried. An alternative search strategy uses topics, a group of predefined terms that share the same concept in any language. For example, the topic “breast cancer” will include results for topics such as “brustkrebs,” which is “breast cancer” in German. Finally multiple queries can be searched concurrently to compare RSF across the terms (comparison) or individually to reflect the RSF of each individual term (individual search), which is more useful when comparing trends across terms in comparison to relative frequency.

Google Trends can also be used to evaluate related queries, which report on related search terms that users also search for alongside the index search terms. “Top” terms represent the most popular search terms scaled to the most commonly searched query as 100, and “rising” terms represent the queries with the biggest increase in search frequency during the requested time period.

On February 6, 2022, Google Trends was queried with keyword terms representing interest in breast cancer screening (“mammogram”) or breast cancer symptoms (“breast pain,” “breast lump,” and “nipple discharge”) as a comparison. “Mammogram” was selected to represent breast cancer screening as this is the standard modality for breast cancer screening. Breast cancer symptoms are common symptoms that patients with breast cancer may experience, or alternatively, patients with otherwise benign conditions may experience and thus require further investigation to rule out a breast malignancy. The 3 most common symptoms were included as search terms. In contrast to “mammogram,” the query “breast self-examination” does not require a medical provider visit and so may be less affected by COVID-19; thus, this was included separately as a search term to be analyzed. A worldwide search from January 1, 2019, to January 1, 2022, using the “all-categories” query category was conducted. January 1, 2019, was chosen as the start date to capture baseline interest, the “worldwide” setting was selected to capture search information worldwide, and the “all-categories” query category was chosen to assess interest in any context and to avoid any bias in filtering search results. Related queries were also retrieved for each of the key terms for the same time period. Further searches to compare peaks of terms were conducted as individual searches based on the results retrieved. Since searches included “how to cure breast lump naturally,” this was compared to the terms “lumpectomy” and “breast biopsy” as possible routine next steps in the management of a breast lump in contrast to natural treatment.

To assess whether a change in search volumes for “mammogram” was significant, we conducted time series forecasting using an autoregressive integrated moving average (ARIMA) model variant model allowing for seasonal variability (seasonal ARIMA). Expected search interest for “mammogram” from January 1, 2019, to January 1, 2022, was estimated using the seasonal ARIMA model based on the searches for “mammogram” during the 5 years prior to COVID-19 (from January 1, 2014, to January 1, 2019)

This study involved cross-sectional analysis of publicly available search engine metadata and does not use data on or involve individual human subjects; thus, it fulfills the criteria for institutional review board exemption.

The purpose of this study was to evaluate the impact of the COVID-19 pandemic on public interest in breast cancer screening and its subsequent recovery. Despite an initial drop in interest in mammography from March to April 2020, it quickly recovered by July 2020. After this period, alongside the recovery of interest in screening, there was a rapid increase in interest for arranging for mammography as indicated by searches for “mammogram near me.”

Previous studies have documented a decrease in cancer screening and diagnosis during the COVID-19 pandemic but not individual’s interest in screening during the crisis [23-25]. The initial dip in search volume could be explained by the postponement of elective visits during the first wave of the pandemic, as searches are usually prompted by upcoming visits and discussions with providers [23,24], or a reluctance in seeking medical attention due to the fear of contracting COVID-19 in the health care setting. As a result of reduced health care contact, newly diagnosed cancer rates declined by 46.4% after the start of the pandemic [26].

Despite the drop in screening, we demonstrated consistent levels of interest in breast cancer–related symptoms and breast self-examination during this period. During times of reduced health care contact, patients continued to use the internet to search for their symptoms. This is concerning given that the use of “Dr. Google” has been linked to increased self-medication and the decision not to see a medical professional [27]. This is reflected in our results showing increased interest in natural treatment of breast lumps, which remains sustained despite the recovery of interest in mammography and breast biopsy.

Our study shows a correlation between health campaign (BCAM) and search behaviors. This result supports previous studies that suggest that infoveillance can measure the success of a campaign in driving information-seeking behaviors in a population [28,29].

There are a few limitations to our research. First, there is a potential overrepresentation of younger, more technologically savvy internet users actively searching for health-related terms. Second, data from Google Trends may not represent a portion of the public who do not have internet access, for example, in countries lacking the infrastructure and technology or with lower socioeconomic status. Third, there may be an overrepresentation of English-speaking users who tend to use Google as a primary search engine. For example, instead of Google, Baidu is the top search engine in China with an 83.46% market share [30]. Lastly, even though results indicate general trends, it does not highlight specific subgroups or give information about the demographics of users who could have a higher share in search volumes.

Similar prior studies have previously documented a decrease in individuals’ interest in screening during the first peak of the COVID-19 pandemic [7]. We have demonstrated that there does not seem to be lasting adverse effects on public interest in breast cancer screening as interest in mammography and arranging for mammograms returned to or exceeded the pre–COVID-19 level. This recovery may have been aided by BCAM. There are similar levels of peak interest in mammography and breast self-examination during the October BCAM despite the ongoing pandemic.

In conclusion, online search interest in breast cancer screening experienced a sharp decline at the beginning of the COVID-19 pandemic with a subsequent return to baseline interest in arranging for mammography following this short period of decreased interest.

Our study shows that despite concerns about the impact of the COVID-19 pandemic on breast cancer screening, interest in mammography quickly recovered. This has implications for health care providers leveraging this recovery to encourage more individuals to get screened, especially among those who may have delayed their mammogram due to the pandemic and for health service resource allocation to respond to this rapid recovery in interest.

Additionally, the study highlights the importance of monitoring changes in search behaviors related to health care during a crisis, as it may reflect changes in health care–seeking behaviors in the general public.

Future work could investigate whether the pandemic had a differential impact on cancer screening rates and outcomes among different populations, including racial and ethnic minority groups, rural populations, and low-income individuals. It would also be important to assess if the COVID-19 pandemic had a long-term impact on cancer outcomes including delayed diagnoses and increased morbidity and mortality.