Seasonal and Geographic Patterns in Tanning Using Real-Time Data From Google Trends

The incidence of skin cancer is increasing, partly due to the usage of tanning beds.^1-3 Few studies have considered the temporal pattern of tanning, with most of the previous studies using cross-sectional data from surveys that are prone to bias of memory recall by individual participants. Herein, we demonstrate a novel methodology and use Internet search data to elucidate seasonality and peak interest times in tanning practices.

The data in this study were collected from Google Trends.⁴ For a given search term, Google Trends computes how many searches have been done relative to the total number of searches done on Google.

We investigated the seasonality of interest in using artificial tanning using the search term tanning salon. Other search terms, such as indoor tanning, tanning booth, tanning solarium, and tanning bed, were considered and analyzed. Because these terms resulted in trends similar to those for the term tanning salon, the analyses for them were not included.

The term UV-free tanning, a type that may also occur at tanning salons, did not yield sufficient search results, and therefore this phrase was not analyzed. A more general term, tanning, was chosen to capture a general interest in tanning.

Normalized search volumes (NSVs) were used from January 2004 to December 2013 for both tanning and tanning salon for Canada, United States, and Australia. Within a given country, time series, mean ratios, seasonal plots, and cosinor analysis were used to identify peak interest times and to test the hypothesis that there was significant seasonal variation in normalized search volume over time. R software, version 3.0.2, was used for statistical analysis. Ethics approval was waived per the University of British Columbia’s ethics review board guidelines.

The interest in tanning and tanning salons was highest in March in Canada and United States, and September for Australia (P = .03). There was a 6-month phase shift for the interest peaks between Australian and the North American counterparts (Figure). Cosinor analysis confirmed this finding, with statistically significant seasonal effects found for tanning salon and tanning for all 3 countries (Table). There was a similar overall trend present with a significant decrease in interest in tanning salons from 2010 onward.

Our results demonstrate that interest in tanning peaks prior to summer months, with the highest interest being during the month of March in Canada and the United States. This could be due to variety of reasons, including preparation of a “base tan” prior to embarking on a vacation during March (spring) break or prior to the summer months. Currently in Canada, tanning safety educational campaigns begin in May or June of each year during the beginning of the summer. Considering our findings, having educational campaigns earlier during the year may help optimize harm reduction and prevention of excessive UV radiation exposure.

According to our findings, the interest in tanning salons and indirectly the use of tanning beds has been declining in Canada since 2010, and in United States and Australia since 2012. New regulation banning indoor tanning⁵ could be contributing to the decreasing trend observed. Although this is promising, further studies are needed to confirm these findings.

There are few limitations associated with the methodology introduced in this study. Age may influence an individual’s choice to use the Internet as a source for information. This, however, may not be of significance in this study, because tanning and use of tanning beds are typically popular among the Internet-savvy young individuals.¹ Although Google Trends does not capture the characteristics of the individuals entering search queries, it has been shown that Internet queries are an accurate proxy for real-life behavior.⁶

Internet search data revealed interest in tanning practices to be seasonal, with peak times in March in Canada and the United States. Our results enable the design of timely harm reduction interventions. Our study introduces a novel methodology that may be used to contribute to research in the field of dermatology. Further studies are needed to confirm these findings.

Corresponding Author: Bez Toosi, MD, Dermatology and Skin Science, University of British Columbia, 835 W 10th Ave, Vancouver, BC V5Z 4E8, Canada (toosi@alumni.ubc.ca).

Accepted for Publication: August 16, 2015.

Published Online: December 30, 2015. doi:10.1001/jamadermatol.2015.3008.

Author Contributions: Both authors had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Both authors.

Acquisition, analysis, or interpretation of data: Both authors.

Drafting of the manuscript: Both authors.

Critical revision of the manuscript for important intellectual content: Kalia.

Statistical analysis: Both authors.

Administrative, technical, or material support: Both authors.

Study supervision: Both authors.

Conflict of Interest Disclosures: None reported.