#PlasticsTwitter: The Use of Twitter Data as a Tool for Evaluating Public Interest in Cosmetic Surgery Procedures Free

Abstract

Background

Plastic surgeons are increasingly turning to social media to market their services. The newly released Twitter Academic Research Product Track (TARPT) database provides free, customizable analysis of keywords that are included in tweets on the Twitter platform. The TARPT tool may provide valuable insight into public interest in cosmetic surgery procedures.

Objectives

The aim of this study was to determine TARPT’s utility in tracking and predicting public interest in cosmetic surgery procedures and to examine temporal trends in tweets related to cosmetic facial and body procedures.

Methods

The TARPT tool was used to calculate the total number of tweets containing keywords related to 10 facial cosmetic procedures and 7 cosmetic body procedures from 2010 to 2020. Annual volumes for respective procedures were obtained from annual statistics reports of The Aesthetic Society from 2010 to 2020. Tweet volumes and procedure volumes were compared by univariate linear regression, taking P < 0.05 as the cutoff for significance.

Results

Variations in tweet volume were observed. Univariate linear regression analysis demonstrated statistically significant positive correlations between tweet volumes and procedure volumes for 7 search terms: “eyelid lift,” “facelift,” “lip injections,” “mastopexy,” “butt lift,” “butt implants,” and “liposuction.” Many procedure-related keywords were not significant, demonstrating the importance of careful selection of Twitter search terms.

Conclusions

The TARPT database represents a promising novel source of information for plastic surgeons, with the potential to inform marketing and advertising decisions for emerging trends in plastic surgery interest before these patterns become apparent in surgical or clinical volumes.

See the Commentary on this article here.

In 2020 alone, there were more than 2,000,000 cosmetic surgical procedures performed in the United States.¹ Despite the COVID-19 pandemic limiting cosmetic surgery availability for several months of the calendar year, public interest in aesthetic procedures remained strong, with more than 200,000 surgical cases performed by The Aesthetic Society members for 4 different procedures: eyelid surgery (blepharoplasty), facelift (rhytidectomy), liposuction, and nose reshaping (rhinoplasty).¹ In 2019, prior to the pandemic, more than 60% of plastic surgeons reported that they used social media to advertise their services, and approximately 80% of plastic surgeons who described their work as primarily cosmetic reported maintaining active professional social media accounts.^2-5 The pandemic has further accelerated the growth of social media in plastic surgery, with many surgeons turning to various marketing platforms such as Twitter (San Francisco, CA), Facebook (Meta, Menlo Park, CA), and Instagram (Meta) to advertise to potential patients.^5,6 Additionally, increased use of videoconferencing platforms during the pandemic has been linked to greater patient interest in obtaining aesthetic procedures to alleviate appearance concerns (particularly above the neck) that are more noticeable on videoconferencing platforms.⁷

As patients continue to use social media and other online platforms to seek information about cosmetic procedures, surgeons seeking up-to-date information on which procedures are popular among patients can track social media trends on public discourse platforms such as Twitter. Twitter is a “microblogging” platform that allows users to post short messages, known as “tweets,” in 280 characters (only 140 characters until 2017) or less.⁸ After a tweet is posted, other users may “like,” “favorite,” or “retweet” the original tweet, which all increase the tweet’s visibility on the platform. Twitter has more than 73 million daily users in the United States who produce more than 400 million tweets per day. Additionally, Twitter appeals to an older audience than other popular social media platforms, such as Snapchat (Santa Monica, CA) and Instagram, with more than 63% of users between the ages of 35 and 65 years, which is particularly appealing to plastic surgeons hoping to market their services to a middle-aged patient population.⁹

Given Twitter’s increasing popularity and the demographics of its users, data regarding Twitter keyword trends theoretically offer great utility to surgeons hoping to gauge the public’s interest in cosmetic surgical procedures. However, although analyses from other search engines such as Google (Mountain View, CA) have suggested that online search data for surgical procedures do correlate with the actual number of procedures performed, there is a paucity of information regarding whether this relationship holds when examining the correlation between tweet volumes related to a given procedure and actual cosmetic surgery procedure volumes.^10-15 Until recently, the method used to gather information about tweet volumes related to a given procedure was arduous and nonintuitive.

In 2021, Twitter launched a new feature called the Twitter Academic Research Product Track (TARPT) database. The TARPT tool is free for academic researchers, intuitive, and offers access to the full archive of tweets since Twitter’s inception in 2006.¹⁶ TARPT offers users an opportunity to gain insight about many different Twitter metrics, including analysis of individual tweets, user lookup, follows, blocks, mutes, likes, and more.

With the TARPT database, the user simply enters the search term of interest (eg, “blepharoplasty”) into the tool, specifies time constraints of interest (eg, 2010-2015), and specifies a geographic location of interest (eg, United States) before executing the search. The TARPT database will then provide a breakdown of the number of tweets featuring the search term per day within the specified time period.

Given the increased use of Twitter in plastic surgery,^4-6 we believe that the TARPT database represents an informative and accessible tool that could be used for allocating surgeon resources and gauging patient interest in plastic surgery procedures. However, due to its recent launch, to the best of our knowledge there are no existing published manuscripts utilizing the TARPT database to examine the relation between Twitter data and plastic surgery procedures performed. As such, this study aimed to evaluate the correlations between cosmetic surgery–related tweet volumes and The Aesthetic Society cosmetic surgery procedure volumes. Describing these correlations and the potential utility of the TARPT tool to gauge public interest in cosmetic procedures may assist plastic surgeons in evaluating, analyzing, and responding to the changing needs of their patients.

METHODS

TARPT

The “Full Archive Tweet Count” component of the TARPT tool was used to evaluate trends in tweet volume over time. Tweet analyses were customizable by search term, time period, and geographic location. After a given search term was entered into the Full Archive Tweet Count component of the TARPT tool and the appropriate temporal and geographic parameters were specified, a report was generated indicating how frequently the search term or keyword appeared in a tweet over the specified time period. For our study parameters, we generated a database of daily tweet totals in the United States for all search terms of interest from January 1, 2010 to December 31, 2020. Tweet volumes for all years after 2010 (2011-2020) were adjusted to account for natural growth in Twitter users and number of tweets per day. Tweet volumes provided by the TARPT tool for each year after 2010 were divided by a factor equivalent to the ratio of the total number of tweets on the Twitter platform during the respective calendar year divided by the total number of tweets in 2010 to standardize tweet volumes over the study period. Adjusted tweet volumes were subsequently used in our statistical analyses.

Search Term Selection

Three prior studies evaluating Google search trends for facial cosmetic procedures, cosmetic body procedures, and breast procedures were used when creating a list of cosmetic procedures (and their affiliated search terms) to be included in our study.^10-12 In addition to these prior studies, other search terms included in this analyses were generated with the “related queries” feature of the Google Trends tool to determine what search terms people most frequently use to search for information related to various cosmetic surgery procedures.¹⁷ For each procedure selected, the search terms or keywords generated included both technical and colloquial terms. Tweet data were collected for search terms and/or keywords related to 10 facial cosmetic procedures and 7 cosmetic body procedures. The 10 facial cosmetic procedures were: blepharoplasty, rhytidectomy, forehead lift, rhinoplasty, malar augmentation, mentoplasty, hair transplant, lip augmentation, lip reduction, and otoplasty. The 7 cosmetic body procedures were: breast lift, breast reduction, breast augmentation, buttock augmentation, buttock implants, liposuction, and abdominoplasty. All search categories and affiliated search terms for facial cosmetic procedures are shown in Table 1 and for cosmetic body procedures in Table 2.

Case Volumes

US case volumes for cosmetic procedures were retrieved from The Aesthetic Society as reported by their available annual statistics reports from 2010 to 2020.¹ We used the years 2010 to 2020 as temporal limits for the study because although Twitter was created in 2006, tweet volumes increased drastically from approximately 2 million annual tweets in 2009 to 35 million annual tweets in 2010, reflecting greater public usage of the platform.⁹

Statistical Analysis

Designating P < 0.05 as indicating significance, we used univariate linear regression to determine the correlation between annual tweet volume for each cosmetic procedure–related search term and the corresponding annual volume of cosmetic procedures performed in the United States according to The Aesthetic Society data. We also provided descriptive statistics on the procedures with the greatest average annual tweet growth over the study period. Finally, we evaluated potential seasonal (spring = March-May; summer = June-August; fall = September-November; winter = December-February) trends in public interest in each procedure by examining monthly tweet volumes for each search term found to be a statistically significant predictor of annual procedure volumes. All statistical and trend analyses were performed with Microsoft Excel version 15.21.1 (Microsoft, Redmond, WA) and SPSS version 26.0.0.1 (SPSS Inc., Chicago, IL).

RESULTS

Correlation of Tweet Volumes and The Aesthetic Society Procedure Volumes

Univariate linear regression analysis demonstrated statistically significant positive correlations between tweet volumes and The Aesthetic Society procedure volumes for 7 search terms, namely: “eyelid lift” (R² = 0.682, P = 0.002), “facelift” (R² = 0.691, P = 0.004), “lip injections” (R² = 0.630, P = 0.011), “mastopexy” (R² = 0.632, P = 0.008), “butt lift” (R² = 0.723, P = 0.001), “butt implants” (R² = 0.699, P = 0.003), and “lipo” (R² = 0.634, P = 0.018) (Tables 3, 4). Several search terms demonstrated no significant correlation between tweet volumes and The Aesthetic Society procedure volumes. Tables 3 and 4 show regression statistics for all search terms associated with facial cosmetic and cosmetic body procedures, respectively.

Greatest Year-Over-Year Average Increase in Tweet Volume

In order to examine trends in tweet volume over the 11-year study period, the percentage change comparing the number of tweets featuring each search term in each successive year of the study period was calculated for each consecutive 2-year sequence (ie, 2010-2011, 2011-2012, etc) from 2010 to 2020. The average annual percentage change for each search term was then calculated, adjusting for the expected increase in tweet volume that would result from a growth of Twitter users. The search terms with the greatest average annual percent increase in tweet volume from 2010 to 2020 included “lip surgery” (+2016.4%), “butt lift” (+1302.4%), “lip injections” (+1094.4%), “nose job” (+699.5%), “eyelid lift” (+500.2%), and “lip implants” (+476.9%) (Figure 1).

Seasonal Trends

Tweets containing the search terms “facelift,” “lip injections,” “mastopexy,” and “butt implants” showed no clear monthly or seasonal trends. Tweets containing the search term “eyelid lift” were most frequent in the spring months (+9.1% increase in tweet volumes compared with average monthly tweet volumes) and summer months (+5.1%) and less frequent in the fall (–3.5%) and winter (–4.3%) months (Figure 2). The same trend was observed for the search terms “butt lift” and “lipo,” with a pattern of increased tweet volumes in the spring and summer months and decreased tweet volumes in the fall and winter months (Figure 2).

DISCUSSION

The results of our study demonstrate that Twitter data may be useful in analyzing trends in public interest related to selective facial cosmetic and cosmetic body procedures. Our analysis detected statistically significant positive correlations between Twitter search terms and US procedure volumes in 7 out of 17 cosmetic procedure categories: “eyelid lift” in the blepharoplasty category, “facelift” in the facelift (rhytidectomy) category, “lip injections” in the lip augmentation category, “mastopexy” in the breast lift category, “butt lift” in the buttock augmentation category, “butt implants” in the butt implants category, and “lipo” in the liposuction category. Many search terms related to the cosmetic procedures studied were not positively correlated with annual procedure volumes, suggesting the importance of careful selection of search terms when attempting to track public interest in cosmetic procedures or when making marketing decisions. With patients’ use of online resources and social media platforms such as Twitter expected to continue to rise in the coming years, Twitter data may represent valuable information regarding trends in patient interest in cosmetic procedures for plastic surgeons and plastic surgery practices.¹⁸

Social media offers an opportunity for the rapid, widespread exchange of information. In the past decade, Twitter has established itself as a leading social media platform, with approximately 40 million daily active users in the United States.⁹ Our results indicate that for select procedures, the amount of public discourse on the Twitter platform (as reflected by the number of tweets containing a cosmetic procedure name or associated search term) was positively correlated with procedure volumes in the United States. Therefore, plastic surgeons may use information obtained from the TARPT database regarding procedures such as blepharoplasty, facelifts, lip augmentation, breast lift, buttock augmentation, buttock implants, and liposuction to guide decisions about marketing and to provide insight about the public discourse surrounding other cosmetic procedures, which can be used to guide expectations for the patient encounter.

The cosmetic procedures with the largest year-over-year average increase in Tweet volume from 2010 to 2020 were “lip surgery” (+2016.6%), “butt lift” (+1302.4%), “lip injections” (+1094.4%), “nose job” (+699.5%), and “eyelid lift” (+500.2%) (Figure 1). These findings align with previous social media research indicating the public’s preference for colloquial rather than technical terms when discussing health information online.^10,19 Interestingly, 1 of the 7 search terms in our study found to have a positive correlation with procedure volumes, “mastopexy,” is more technical than colloquial. Studies by both Tijerina et al¹¹ and Wilson et al²⁰ proposed that technical terms may be of more utility than colloquial terms when analyzing public interest in breast augmentation, which aligns with our findings.

Previous research has shown the impact that high-profile plastic surgery–related media coverage can have on public interest in cosmetic procedures.^10-12,18 Interestingly, the months with the greatest percentage increase in tweets for the search terms “lip injections,” “lip surgery,” and “nose job” were all associated with high-publicity media events. For “lip injections,” tweets spiked during the month of September 2017, which was when social media influencer Kylie Jenner revealed to more than 39 million Twitter followers that she got “fuller lips” on her reality television series “Life of Kylie.” ²¹ Jenner also likely caused a spike in tweets containing the phrase “lip surgery” in October 2018, when she revealed on her Instagram story that she had another lip procedure to create a more “natural lip shape.” ²¹ For “nose job,” tweets spiked in February 2018, which was when television star Naomi Olindo revealed on her Instagram account that she had acquired a “new nose,” resulting in news articles published by high-profile media outlets such as People, E!, and BravoTV.²² Previously, for some plastic surgery procedures, increased public interest on social media in response to high-profile media announcements has been associated with an increase in the actual number of procedures performed in the United States.^10,23 As such, for plastic surgeons, the TARPT database represents an opportunity to track public interest in cosmetic procedures in response to high-profile media coverage, which can be used to inform marketing strategies.

Although there were significant positive correlations between tweet volumes and procedure volumes for several of the search terms included in our study, there were also search terms with tweet volumes and procedure volumes that were not correlated. This suggests the importance of careful search term selection when utilizing the TARPT database as a barometer of public interest for cosmetic procedures. Interestingly, when comparing the TARPT database vs the Google Trends database as a means to measure of public interest in cosmetic procedures by searching for the same terms, Google searches were more predictive of procedure volumes than tweet volumes. Tijerina et al reported a greater percentage of significant positive correlations between cosmetic procedure–related Google searches and American Society of Plastic Surgeons (ASPS) case volumes than we discovered in our analysis.^10-12 However, the stronger associations between Google searches and procedure volumes when compared with our study is likely due to the varying natures of the Twitter social media platform and the Google search engine. The Twitter platform mimics dialogue between users and is more colloquial, whereas Google is a search engine where patients are more likely to enter or be linked to technical names to gain information about a procedure or a provider for a particular procedure. Positive associations between tweet volumes and procedure volumes for more colloquial terms, such as “eyelid lift,” “butt lift,” and “lipo” suggest that the use of more colloquial terms to describe procedures may result in even stronger associations between Twitter data and procedure volumes, which could minimize the difference in predictive ability we observed between the Google Trends and TARPT tools.

A limitation of the TARPT tool is that a tweet must exactly replicate the search term of interest for it to be picked up by the tool. If a tweet contains a slight variation of the search term of interest (eg, the phrase “lip reduction procedure” vs “lip reduction surgery”) it would not be picked up by the tool.

Although Twitter’s primary use is as an interactive discussion tool rather than a strict information source, this does not mean that it lacks relevance to the plastic surgery community. Several recent articles have discussed the tremendous impact of social media in the field, with one article claiming that plastic surgeons lacking digital communication and social networking skills may lose patients to more tech-savvy surgeons.^24-26 Another called for plastic surgeons to become engaged with social networking tools such as Twitter to better educate the public and to help to combat the massive amounts of misinformation that are often provided to patients online, which can result in poor decision-making.²⁷ Plastic surgeons have taken notice of the calls for increased social media presence, with rising social media usage within the field according to a survey administered to ASPS members.⁴ The TARPT database represents an opportunity for increasingly social media–savvy plastic surgeons to receive real-time information about public interest in cosmetic surgery procedures, which can be used to assist in marketing and resource allocation decisions.

Recent evidence indicates that investments in social media websites such as Facebook and Instagram offer more return on investment than Google search engine optimization and referral websites such as Yelp and RealSelf for plastic surgeons hoping to grow their practice.²⁸ The Twitter platform represents an ideal opportunity for plastic surgeons to market to potential patients, in part because current Twitter discussions about plastic surgery procedures are dominated by the public, not by practicing plastic surgeons. A recent analysis of nearly 3000 tweets containing the words “plastic surgery” and/or the hashtag #plasticsurgery showed that more than 70% of tweets analyzed were authored by members of the public, whereas only 6.0% of tweets were by written by board-certified plastic surgeons.²⁷ Another study revealed that the majority of Instagram top posts associated with common plastic surgery hashtags (such as #facelift, #rhinoplasty, #tummytuck) were from surgeons based outside the United States, indicating the potential for plastic surgeons within the United States to improve their social media presence.²⁹ Currently, nearly half of all plastic surgery patients use social media when choosing their doctor.¹⁸ Our results indicate that for certain plastic surgery procedures, increased Twitter discussion correlates to increased procedure volumes. A tech-savvy plastic surgeon has the potential to discover what patients are talking about online, engage in the Twitter conversation, and massively expand their patient outreach.

There are several limitations to our study. First, although we measured the quantity of tweets containing specific search terms related to many cosmetic procedures, the contextual content of the tweets was not evaluated. As such, we cannot determine whether tweets containing the search term of interest endorsed or objected to the procedure of interest. However, the goal of this study was to evaluate the presence or absence of a relationship between the popularity of a given search term on Twitter and the number of procedures performed, and as such, we believe our conclusions remain valid. Another limitation stems from the case volumes that we received from The Aesthetic Society annual reports. Not all cosmetic procedures are reported by The Aesthetic Society individually during a given year, and this deficiency has the potential to influence the correlations observed. Next, the search terms utilized in this study were largely derived from 3 previous studies evaluating public interest in various cosmetic surgery procedures.^10-12 We recognize that some search terms used in the previous studies were technical in nature, and there could be other, more colloquial terms related to procedures whose Twitter data may better predict procedure volumes. Although we employed the “related queries” feature of the Google Trends tool to investigate other potential search terms, we recognize that our study may not have captured all search terms (both technical and colloquial) related to a procedure in our study. Further dedicated studies will be necessary to exhaustively characterize the ideal keywords for a given procedure. Finally, the TARPT database provides limited information on the demographic information of users whose tweets are represented in our study. As such, we are unable to assess if the included study population is representative of the US population. However, according to the ASPS, the age range of patients who receive the most cosmetic surgery procedures (30-54 years) aligns with the 2 age groups who represent the greatest number of Twitter users (25-34 and 35-49 years), and therefore it is reasonable to believe the Twitter users whose data are reflected in this study are at least partially representative of cosmetic surgery recipients.^1,30

CONCLUSIONS

Our study demonstrates that for certain facial cosmetic and cosmetic body procedures, tweet volumes containing relevant search terms displayed statistically significant positive correlations with annual case volumes of US cosmetic surgery procedures. We also described seasonal trends related to tweets for various cosmetic procedures, provided insight on the procedures experiencing the most rapid growth in popularity on the Twitter platform, and linked growth on Twitter to high-profile media coverage. This information may prove useful to surgical marketing teams when attempting to devise advertising strategies in the United States. The TARPT database represents a free, easy to navigate, and largely unknown tool that has the potential to serve as an extremely powerful information source for surgeons interested in learning about which procedures patients are actively discussing online, especially if the use of Twitter in plastic surgery continues to grow at the rate that we have observed in the past decade.¹⁸ Tracking trends in Twitter engagement for various procedures can help surgeons to tailor their marketing strategies and to anticipate patient inquiries during the patient encounter, which can lead to the creation of decision aids and other tools that may result in more informed patient decision-making.

Disclosures

The authors declared no potential conflicts of interest with respect to the research, authorship, and publication of this article.

Funding

The authors received no financial support for the research, authorship, and publication of this article.

References

Author notes