Abstract
Background/Aim: Google Trends (Google Inc., USA) is a web-based service that offers data on the time and volume of Google Internet searches. We used Google Trends to identify increases in search traffic caused by cancer campaigns, which are held in October and November each year for breast cancer (BC) and prostate cancer (PC), respectively, in our region. We investigated whether this increase in search traffic coincided with increased research funding or public donations. Materials and Methods: Google searches for PC were compared and scaled against searches for BC from January 2010 to December 2021. The month with the highest volume of search hits for either term was given a relative search volume index of 100. All other months for both search terms were scaled against this index of 100. The degree of government research spending and public donations were contrasted against search traffic. Results: Search volume increased for both campaign months (49% for BC in October, 5% for PC in November). The mean relative search volume for each campaign’s specific month was 72.6 for BC, 18.4 for PC; mean difference of 50.75 (95% confidence interval=42.7-58.8, p<0.001). Annual mean (±standard deviation) expenditure of Government research grants was $22.2 million (± $1.5 million) for BC and $9.7 million (± $2.9 million) for PC with a mean difference of $12.5 million (95% confidence interval=$10.5-14.6 million, p<0.001). Conclusion: The BC campaign resulted in a statistically significant increase in search traffic compared to that for PC and correlated with statistically significant increases in research spending. Google Trends offers a method of evaluating cancer campaign efficacy and public interest.
Cancer awareness and fundraising campaigns have become important tools to assist multiple stakeholders in the community including researchers and ultimately patients. Increasingly, members of the public are using the Internet and social media to gather information about health-related topics. Google Trends (GT; Google Inc., USA, www.google.com.au/trends) is a web-based service that offers data on the time and volume of search terms for any topic. It has been used for numerous applications in healthcare as varied as assessing vaccine hesitancy during the COVID-19 pandemic and assessing the seasonality of multiple sclerosis (1, 2). Observation of search traffic can be used to quantify public interest in health conditions and may be used to evaluate the success of campaigns aimed at raising awareness of a health condition, with potential translation to research funding.
Two well-known cancer campaign fundraisers are the Pink Ribbon (breast cancer) campaign held in October and the Movember (prostate cancer) campaign, held in November. The Pink Ribbon campaign directs its focus primarily towards breast cancer, but also towards gynaecological cancer in women. The Movember campaign is directed primarily towards prostate cancer, but also promotes awareness of testicular cancer and psychological health in men. The aims of these campaigns are to enhance public awareness of the conditions, to raise funds from public donations and to promote research and development for better treatment options.
Evaluation of a cancer campaign’s success is important for two reasons. Firstly, identifying which campaigns are failing to achieve their goals enables the organiser to alter the campaign’s approach, with new strategies implemented to improve their efficacy. Secondly, understanding which cancer campaigns are successful enables the organiser to look closer at which features helped promote their success and implement them more precisely into future campaign endeavours. The ultimate goal is to improve public awareness and patient outcomes. One such metric of success is the volume of search engine traffic, as this reflects how frequently members of the public are seeking out information on the condition. Other metrics of campaign success include the amount of public donations, as well as government funding invested in research and development, as this is frequently a reflection of public engagement.
In this study, we tested our hypothesis that despite prostate cancer posing a higher burden of disease, breast cancer funding and public donations remain higher and this can be attributed to greater public interest in the disease and greater cancer campaign efficacy as reflected by higher number of Internet searches during campaign months. We used GT to reflect how much search traffic increased during cancer campaign-specific months. We analysed the amount of National Health and Medicine Research Council (NHMRC) grants awarded to fund research and development for both cancer types, as well as the volume of public donations received by the major cancer foundations for breast and prostate cancer. We contrast funding achieved with the burden of disease posed by breast cancer and prostate cancer in Australia.
Materials and Methods
Internet searches. Google searches for ‘prostate cancer’ were compared and scaled against searches for ‘breast cancer’ on a monthly basis from January 2010 through to December 2021 using GT. The highest volume of search hits was for breast cancer in October of 2010. That month’s search traffic was awarded a relative search volume (RSV) index of 100. All other months for both search terms were compared and scaled against the volume of searches made on this date. For example, an RSV of 40 would indicate that 40% as many searches were observed for a particular month when compared to the volume of searches for breast cancer in October 2010.
Cancer and fundraising statistics. NHMRC research grants awarded towards research and development are made publicly available on request (https://www.nhmrc.gov.au). We acquired these figures for breast and prostate cancer from 2010 to 2020 and compared fundings. Annual reports examining revenue from donations and fundraisers for the major charities responsible for contributing funds to breast and prostate cancer were reviewed. These charities included the National Breast Cancer foundation, Breast Cancer Network Australia, Movember Foundation and Prostate Cancer Foundation Australia (3-5). Publicly released cancer statistics pertaining to the burden of disease for breast and prostate cancer, including incidence rates, mortality rates and public expenditure, were accessed from government website reports (6).
Statistical analysis. Mean RSVs for breast and prostate cancer were calculated for the total 132 months. The comparative increase in search traffic for the campaign specific months was compared to the other months of the year. The RSVs for breast cancer during the months of October from 2010 to 2021 were compared to the RSVs for prostate cancer during the months of November from 2010-2021. Yearly NHMRC grants awarded to support research into breast cancer were compared to those for prostate cancer for the same time period. Statistical analysis was conducted using IBM SPSS statistics versions 23 (IBM, Armonk, NY, USA). Independent-sample t-testing was performed for comparisons of RSVs and annual research funding (in Australian dollars). Values are quoted as the mean±standard deviation. A p-value of less than 0.05 was considered statistically significant. Normality of data was tested using the Shapiro–Wilk test.
Results
Search traffic. The GT search figures from January 2010 to December 2021 reveal peaks in searches for breast cancer during October of each year; the peaks are particularly prominent until October 2015, at which time they level out. Further peaks can also be noted in May, which are likely to be related to the annual breast cancer ‘Mother’s Day Classic’ walk, held on the 14th of May each year (Figure 1). No obvious yearly peak search months can be identified for prostate cancer. Search hits for both cancer types show a trend towards decreasing during the months of December and January. A single peak in prostate cancer searches can be seen for July of 2013, when a large volume of media outlets reported on the SELECT trial, which memorably suggested a link between omega-3 fatty acid consumption and an increased risk of prostate cancer (7). Interestingly, search traffic remained relatively stable during the COVID-19 pandemic when compared to pre-pandemic levels.
Graph comparing the relative search volume for breast cancer and prostate cancer by month for 2010 to 2021.
The mean yearly RSVs for breast cancer and prostate cancer from 2010 to 2021 were 50.6±3.72 and 20.5±1.80, respectively, revealing that breast cancer received on average 240% more search traffic than prostate cancer each year. The mean RSV for each cancer’s campaign-specific month was 72.6±12.59 for breast cancer and 21.6±2.44 for prostate cancer (Figure 2), with a mean difference of 50.75 (95% confidence interval=42.7-58.8, p<0.001). During October, search traffic for breast cancer increased by 49% when compared to other months of the year, whereas during November, search traffic for prostate cancer increased by 5% when compared to other months of the year (Table I).
Annual National Health and Medicine Research Council funding for breast and prostate cancer for 2010 to 2021.
Relative search volume (RSV) data summarised for breast cancer and prostate cancer from data for 2010 to 2021 obtained using Google Trends.
Government spending. Between 2010 and 2021, the mean annual NHMRC research grant spending on breast cancer was $22.2 million (±$1.5 million) versus $9.7 million (±$2.9 million) for prostate cancer. Independent t-testing revealed a mean difference of $12.5 million (95% confidence interval=$10.5-14.6 million, p<0.001) (Figure 3).
Relative search volume for breast cancer and prostate cancer during their campaign months, October and November, respectively, for 2010 to 2021.
Public donations. Public donations were marginally higher for prostate cancer, with the National Breast Cancer foundation raising $128 million between 2015 and 2020, and the BCNA raising $36.4 million, a total of $164.4 million. The Movember Foundation raised $133.3 million in Australia and Prostate Cancer Foundation Australia raised $64.9 million in the same period; overall totalling $198 million. There was a general trend toward increased donations over the time period for the prostate cancer charities, with the donations for breast cancer remaining relatively stable (Figure 4) (3-5, 8).
Revenue generated through donations by public charities; National Breast Cancer Foundation (NBCF), Breast Cancer Network Australia (BCNA), Prostate Cancer Foundation Australia (PCFA) and Movember Foundation (MF).
Cancer morbidity and cost. In 2018, the age-standardized incidence rate for breast cancer in Australia was 75 cases per 100,000 persons and for prostate cancer was 176 cases per 100,000 males (Figure 4). In 2014, the total number of deaths related to prostate cancer was 3,309, compared to 3,012 for breast cancer (9). Public expenditure for breast cancer was higher than that for prostate cancer (10) at 4% versus 6.7% (Figure 4).
Discussion
Increasingly, patients are resorting to web-based searches and social media for health information (10, 11). One study looking at oncological outpatients found that despite 90% of the study participants feeling they had been given adequate information from their treating team, over half still sought out additional information, most of whom resorted to the Internet (12). GT identifies obvious spikes in searches for breast cancer every October, in keeping with the timing of the pink ribbon campaign. Unlike the pink ribbon campaign, spikes in search terms for prostate cancer were not seen during the Movember months, suggesting the campaign is failing in one of its goals, which is to raise awareness amongst the public. There are numerous factors that could potentially explain this phenomenon.
One such factor which could possibly explain the greater number of searches for breast cancer and prostate cancer is the relative difference in burden of disease. If it were the case that breast cancer has a higher burden of disease, this could provide a potential explanation for its increased search traffic throughout the year and for its campaign’s awareness month. Examining the data shown in Table II; however, it can be seen that prostate cancer affects more individuals, with an age-adjusted incidence of 176 versus 75 per 100,000 persons. It is also responsible for higher mortality, with 3,309 deaths compared to 3,012 deaths attributable to breast cancer in 2014 (9, 13). These figures suggest the increase in search terms for breast cancer is unlikely to be due to the disease having a greater impact on family and friends through diagnoses or deaths of loved ones. In addition, prostate cancer received on average only 43% of the NHMRC research grant dollars that were spent on breast cancer, with this figure dropping to as low as 28% in 2021. This is a disproportionate sum considering that the number of disability-adjusted life years attributable to prostate cancer is around 70% of those attributable to breast cancer (14).
Comparison of incidence and public expenditure on breast and prostate cancer for the 2020-2021 financial year in addition to mortality for 2014.
Public donations were marginally higher for prostate cancer than for breast cancer. These data become muddied when calculating exact amounts, as the Movember Foundation donates some of its proceeds directly to Prostate Cancer Foundation Australia. Despite the lack of increase in search volume during the Movember campaign, the increase in revenue year-on-year and the large number of donations to the Movember foundation is indicative of considerable success in that domain.
The varying degrees of public interest in these two types of cancer are reflected by government investment. Cancer Australia (www.canceraustralia.gov.au) is the Australian government’s national cancer control agency. Their mission statement is to “…strengthen and provide advice on the Australian Government’s strategic focus on cancer control and care”. They release annual reports on how grants and funding have been allocated. In their 2019-2020 annual report, the term ‘breast cancer’ was written 42 times, in comparison to ‘prostate cancer’, which was written once. For comparison, the next most frequently mentioned cancer was cancer of the lung, which was mentioned 29 times. Comparison was made with the 2012-2013 annual report selected at random which revealed (excluding abbreviations and indexes) breast cancer mentioned 79 times versus a mere 15 mentions of prostate cancer (15). Whilst this is a somewhat rudimentary method for evaluating the agency’s priorities, it is nonetheless interesting and in keeping with the findings of this study.
Examining GT graphs has revealed other interesting peaks and troughs in search traffic. The lowest volume of searches for breast and prostate cancer were consistently seen during the months of December and January, a phenomenon that has been observed globally for all cancer types and is suspected to be due to a tendency towards disregarding health issues over the holiday months (16). Furthermore, media headlines which pay attention to journal articles yield significant increases in search traffic, as was observed by the promulgation of the SELECT trial in July 2013.
This study has some limitations. The evaluation of search traffic was limited to a single Internet search engine. Other key terms may have had increased search traffic, such as ‘Movember’, instead of ‘prostate cancer’, and the impact alternative search terms has on public awareness of the condition was not evaluated. Furthermore, decreasing searches for prostate cancer may be an indication of pre-existing public understanding of the condition. This would preclude the need for increased Internet search traffic during campaign months and would suggest prior partial campaign success.
Conclusion
This study suggests that GT offers a useful tool for evaluating cancer campaign efficacy and public interest in disease. Despite the higher burden of disease posed by prostate cancer, search traffic for breast cancer was significantly higher during its cancer campaign months. This coincided with significantly larger NHMRC research grants and a trend towards higher donations to not-for-profit cancer foundations. Techniques that have been employed by the pink ribbon campaign should be analysed and implemented into the Movember campaign, as this may result in greater public awareness, cancer detection and funding for research and development. Observing GT also reflects popular interest in media articles as evidenced by the SELECT trial in July of 2013 and can be used to give insight into times of the year when society is less receptive to information regarding health and wellness, such as during the holiday season in December.
Footnotes
Authors’ Contributions
JT: Writing, data analysis and conceptualization. JRMC: Writing, editing and data analysis. AD: Data curation and writing. NL: Conceptualization, supervision.
Conflicts of Interest
The Authors have no conflicts of interest to declare.
- Received November 18, 2022.
- Revision received December 5, 2022.
- Accepted December 6, 2022.
- Copyright © 2023 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
