Quality evaluation of internet resources related to breast surgery in major search engines in Korea

Article information

Arch Aesthetic Plast Surg. 2024;30(2):44-52

Publication date (electronic) : 2024 April 30

doi : https://doi.org/10.14730/aaps.2023.01011

Hee Jong Lee¹, Dong Jin Kim², Nara Lee², Jin Sup Eom², Tae Suk Oh², Eun Key Kim^,2

¹DABIDA Plastic Surgery Clinic, Seoul, Korea

²Department of Plastic Surgery, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea

Correspondence: Eun Key Kim Department of Plastic Surgery, Asan Medical Center, University of Ulsan College of Medicine, 88 Olympic-ro 43-gil, Songpa-gu, Seoul 05505, Korea E-mail: nicekek@korea.com

Received 2023 July 5; Revised 2023 December 14; Accepted 2024 April 20.

Abstract

INTRODUCTION

As of 2019, over 4.5 billion people, or 58.8% of the global population, were using the internet, with an average daily usage of 6.5 hours per person [1-3]. When seeking health information, individuals typically access portal sites, which are the initial sites they visit through their internet connection. In Korea, as of 2014, the market shares were distributed as follows: Naver at 73.0%, Google at 14.5%, and Daum at 12.3% [4]. By the first half of 2022, the figures had shifted to Naver holding 60.1%, Google increasing to 29.2%, and Daum dropping to 5.2% [5]. Although Naver remains the dominant portal, Google has seen a substantial increase in its share over the past eight years, whereas Daum’s share has declined to a single digit (source: www.internettrend.co.kr). On Google’s search engine, approximately 7% of queries are for health-related information, with over 1 billion searches for health-related topics occurring daily.

Internet searches can induce positive changes in health behavior by providing patients with health-related information; however, it is challenging for patients to assess the quality of medical information [6,7]. Additionally, understanding the nature and quality of online information can be beneficial for medical personnel who interact with patients. It is essential to evaluate the information that patients currently find online. Therefore, this study aimed to assess the quality of websites and the information they provide on “gaseum” (breast)/“yubang” (breast) implant surgery. This topic was chosen as it is a primary search term for patients interested in cosmetic surgery or those with breast cancer seeking breast reconstruction. The evaluation was conducted using three major portal sites commonly used in Korea and employed systematic health information evaluation standards.

METHODS

Primary search terms and primary search term groups

The search strategy included both main and adjunct search terms. The primary search terms were “gaseum” (breast) implant and “yubang” (breast) implant. The adjunct search terms are listed in Table 1. In the dictionary, “gaseum” is defined as the front part of the body between the stomach and neck, whereas “yubang” refers to the functional reproductive organs that include mammary glands, ducts, and fatty tissue. “Gaseum” is commonly used in everyday language, while “yubang” is more prevalent in academic and medical contexts. As in other languages, these synonyms for “breast” carry different connotations and are not entirely interchangeable (see, for instance, “breast” versus “boob,” “bosom,” and other colloquial terms in English). A search term was formulated by combining the two main primary search terms with 15 adjunct search terms, resulting in a total of 30-word combinations. The adjunct search terms were organized into six smaller groups based on their relevance (Table 1).

Table 1.

Combinations of the adjunct search terms

Study subject

In July 2022, the search engines Google (http://google.co.kr), Naver (http://www.naver.com), and Daum (http://www.daum.net) were utilized for research. All searches were conducted in Korean. Prior to searching, all search records and cookies were deleted, and searches were performed using the Google Chrome browser. To facilitate understanding, the data collection and analysis process is illustrated in a flowchart in Fig. 1. The top 20 websites resulting from the search terms were assessed. On Google, these top 20 websites were evaluated as they appeared. In contrast, Naver and Daum included their own content (i.e., blog, cafe, and knowledge IN) in the search results; thus, the top 20 websites from these integrated searches were also evaluated. The number of significant websites identified for each term was assessed, and the distribution and frequency of categories, based on the character (form and provider) of the websites, were analyzed. Furthermore, the format and content of the websites were evaluated and scored using the Korean Medical Association (KMA)’s Internet Health Information Certification Standards, which were developed by the KMA in 2003. For ease of reference, this score will be termed the “KMA-S score.” Our Institutional Review Board granted exempt status for this protocol (approval number: 2023-0128).

Fig. 1.

Flowchart of the process of data collection and analysis.

Website categories and subgroups

The resulting websites were classified into 19 categories (Table 2), and each category was classified into six subgroups according to website type and five subgroups according to the provider (Table 3).

Table 2.

Website categories

Table 3.

Category subgroups (form and provider)

RESULTS

Number of significant websites

In the analysis of search results for selected primary search term combinations, 1,800 entries were examined, of which 873 websites (48.5%) were deemed insignificant. The search results for 30 Korean terms, which combined the two primary search terms with 15 adjunct search terms, are presented in Table 4. This table also compares the number of significant websites across different search engines. Overall, significant websites accounted for 51.5% of the total. Independent of the primary search terms used, Google consistently provided the highest number of relevant results, a finding that was statistically significant. Regardless of the search engine, the search terms “reconstruction of ‘yubang’ (breast) implant” and “reoperation of ‘gaseum’ (breast) implant” consistently yielded the highest number of significant websites. Conversely, the terms “recommendation for ‘yubang’ (breast) implant” and “review of ‘yubang’ (breast) implant” resulted in the fewest significant websites.

Table 4.

Number of significant websites from the search engines for adjunct search terms

Fig. 2 summarizes the results when the 15 adjunct search terms were classified into six groups. The two adjunct search term groups with the highest number of significant websites were “gaseum (breast) implant–group 5 (removal, reoperation)” and “yubang (breast) implant–group 3 (breast cancer, reconstruction)” (Fig. 2). Significant websites identified in the search results were categorized into 19 categories, while insignificant websites were placed in a 20th category (Supplementary Table 2).

Fig. 2.

Comparison of significant websites according to adjunct search term groups. Group 1 (operation, plastic surgery), group 2 (augmentation), group 3 (breast cancer, reconstruction), group 4 (complication, side effects, rupture, contracture), group 5 (removal, reoperation), group 6 (selection, recommendation, cost, review). A, Gaseum (breast); B, Yubang (breast).

Analysis of the website categories

Category distribution

From the total results, blogs managed by healthcare providers were the most common (15.1%), followed by news in the form of medical personnel advice (12.1%) and medical websites (7.4%). In contrast, social network services (SNS), legal websites, and academic organizations registered the lowest incidences. Among blogs, 80.42% of the 337 analyzed were authored by healthcare providers, while only 5.04% were contributed by patients or patient organizations. In the news category (n=323), columns and advice from medical personnel were the most prevalent, accounting for 67.49%.

Subgroups of website types

When the websites were categorized into six subgroups by type, blogs were the most common (36.4%). News websites accounted for 34.8%, and medical homepages accounted for 22.0% (Fig. 3). Regarding the overall results, in contrast to Google and Naver, the blog group on Daum comprised a significant portion, accounting for 72.7% (P<0.001) (Supplementary Table 3).

Fig. 3.

Distribution of significant websites according to website type.

Subgroups of providers

When the websites were grouped into five subgroups according to the provider, websites by healthcare providers were the most common at 49.5%, followed by third parties at 45.1%. Websites by manufacturers and distributors were the least common, at 1.1%, whereas the public organization group showed a similarly low result of 1.9% (Fig. 4, Supplementary Table 4).

Fig. 4.

Distribution of significant websites according to provider subgroups.

Comparison of average scores between search engines

Search engines

The overall average quality score across all major websites was 13.36. Google achieved a significantly higher average score of 14.04 compared to Naver, which scored 13.22 (P=0.041), and Daum, which scored 12.45 (P<0.001). This was consistent regardless of whether the primary or adjunct search terms were used.

Adjunct search terms (groups)

The search results from each search engine were evaluated using the KMA-S score (Table 5). The average scores across the search engines were then compared and analyzed. The highest scores were observed for the adjunct search term “reconstruction,” followed by “rupture,” while the term “cost” received the lowest score. Significant differences were noted between Google and Daum for the search terms “breast cancer” and “rupture,” and between Naver and Google for the terms “cost” and “review” (P<0.05) (Table 5). When comparing the average scores among search engines according to the adjunct search term group, statistically significant differences (P<0.05) were found between Naver and Dum in group 1 and between Google and Daum in groups 3 and 4 (Fig. 5).

Table 5.

Average quality scores of the search engines for adjunct search terms

Fig. 5.

Comparison of the average scores among the search engines according to adjunct search term groups. Group 1 (operation, plastic surgery), group 2 (augmentation), group 3 (breast cancer, reconstruction), group 4 (complication, side effects, rupture, contracture), group 5 (removal, reoperation), group 6 (selection, recommendation, cost, review). KMA-S, Korean Medical Association (KMA)’s Internet Health Information Certification Standards.

Analysis of the categories

The average scores of the search engines for specific categories were compared (Table 6). The search results of category 17 (YouTube) had the highest average score among the three search engines, whereas the lowest score was found for category 6 (medical forums). The differences between Naver and Google and between Naver and Daum in category 14 (news 1) were statistically significant (P<0.05). Similarly, the average scores of the search engines according to website type subgroups were analyzed (Fig. 6). The highest average scores were found in group 6 (other media), while the lowest average scores were observed in group 1 (homepage). Furthermore, the average scores of the search engines according to provider type subgroups were compared (Fig. 7). The highest average scores were identified for group 2 (public organizations). Statistically significant differences were observed between Google and Daum in group 1 (healthcare providers), as well as between Google and Daum and between Naver and Daum in group 5 (third parties) (P<0.05).

Table 6.

Average quality scores of the search engines for categories

Fig. 6.

Comparison of the average scores among the search engines according to website type. Group 1 (homepages), group 2 (journals/abstracts), group 3 (blogs), group 4 (cafes), group 5 (news), group 6 (other media). KMA-S, Korean Medical Association (KMA)’s Internet Health Information Certification Standards.

Fig. 7.

Comparison of the average scores among the search engines according to provider subgroups. Group 1 (healthcare providers), group 2 (public organizations), group 3 (patient, legal organizations), group 4 (manufacturers and distributors), group 5 (third parties). KMA-S, Korean Medical Association (KMA)’s Internet Health Information Certification Standards.

DISCUSSION

In the era of rapidly advancing internet technology, the ease of access to medical information has significantly improved, underscoring the vital role of digital health literacy [8-12]. This concept encompasses the ability to effectively find, understand, evaluate, and use health-related information from electronic sources to address health issues or make informed health decisions.

A study by Choi et al. [13] revealed that about 66.2% of Koreans turned to internet portals, YouTube, and television for health information. Building upon this, further research focused on the top 20 websites from the most used search engines in Korea, particularly for information on breast augmentation and implants [14]. The effectiveness of information varied significantly based on specific search terms, illustrating the need for linguistic precision when seeking health information online.

The website type and provider subgroups show similar trends, with blogs (36.4%) and healthcare providers (49.5%) receiving the highest scores. Conversely, the highest average scores were found for medical care-related government entities (1.0%), academic organizations (0.1%), journals/abstracts (0.8%), and public organizations (1.9%), which represented only a small fraction of the total. This indicates that blogs and healthcare providers need to focus on improving the quality of the information they provide online. Meanwhile, authoritative bodies such as government agencies and academic organizations, including the Korean Society of Plastic Surgeons and the Korean Society of Aesthetic Plastic Surgeons, should increase their online informational output. While it is crucial to produce high-quality information, the primary goal of a website is to disseminate knowledge. Therefore, it is equally important to consider strategies for making this quality information more visible and accessible.

When comparing the percentage of significant websites, there was no difference between the main keywords (“gaseum” [breast]/ “yubang” [breast]), with each appearing six times among the top 12 search terms with the highest percentage. However, the two keywords with the highest percentage, “yubang” (breast) implant reconstruction and “gaseum” (breast) implant revision surgery, highlight distinct focuses. Yubang is predominantly associated with breast cancer and reconstruction, while gaseum is more linked to cosmetic augmentation and revision surgery. In reconstructive surgery contexts, yubang is more commonly used than gaseum, especially following breast cancer surgery. This terminology is prevalent in university hospitals, where most breast cancer and reconstructive procedures are performed, influencing the language used by doctors and in doctor-patient communications. If search results vary between synonyms based on additional keywords, it is advisable for healthcare providers to choose terms that align with their informational goals, and for users to select terms that meet their information-seeking needs.

The role of healthcare providers as a source of online information was significant, with nearly half of the content originating from this sector. Interestingly, while public organizations contributed a smaller portion of the information, their content was recognized for its high quality. Although one might expect the material provided by healthcare providers to be of the highest quality, the actual significant site results showed no significant difference when compared to those of third parties, at 49.5% and 45.1%, respectively. This is likely due to the prevalent promotional use of these sites. Concerns have been raised regarding the promotional nature of some of these sites.

YouTube has proven to be a particularly reliable source, achieving high scores in areas such as authority, author verification, and scientific soundness. This reliability can be attributed to the direct involvement of medical personnel, transparent account information, and the platform’s capacity to incorporate user feedback and offer regular updates. These findings indicate that healthcare providers should view YouTube as an effective tool for disseminating information.

When comparing search engines, Google was found to provide higher quality health information than local Korean search engines such as Naver and Daum. Google’s algorithm-driven, out-link method was considered less susceptible to distortion, in contrast to the in-link method used by domestic engines, which may include more human intervention and potential bias [15-19].

A previous international study conducted in 2002 also evaluated websites that provided information on plastic surgery. It found that 34% of these websites offered false or misleading information, suggesting that healthcare providers need to assist patients in distinguishing accurate information, as demonstrated in this study [20].

The study had several limitations, one of which was its exclusive focus on Korean websites. This narrow scope does not represent the behavior of users who access foreign websites. Additionally, the study did not comprehensively address the public’s needs and satisfaction levels when seeking medical information. The significant differences in search algorithms on mobile devices, particularly in a country with high smartphone usage like Korea, highlight the necessity of evaluating mobile search results [21,22]. Additionally, the focus of the study on breast implants and related complications may not be fully applicable to other medical conditions.

This research provides meaningful insights into the changing dynamics of digital health literacy and the critical importance of evaluating online health sources carefully. It underscores the necessity for accurate search strategies, particularly in an era where misinformation can spread rapidly. For instance, one could implement a certification system for hospital websites and blogs by government authorities, or improve visibility by regularly publishing credible, academically rigorous content and promoting it through media outlets.

In the search for “gaseum” (breast)/“yubang” (breast) implants, approximately half (51.5%) of the results were relevant to the search term. However, the average score was 13.36, which is significantly lower than the 21 points scored by the journal/abstract category (i.e., the category with the highest score). The significant websites were primarily in the form of blogs or news, with most information provided by medical providers. The score for these sources was 13.16 points, falling below the overall average. Nevertheless, among healthcare providers, blogs managed by medical personnel and YouTube channels featuring medical personnel received high scores.

To improve the quality of medical information available online, it is crucial for information providers, search engines, and public organizations to strive to ensure that the information is of high quality and readily accessible to users. As part of this initiative, the website will feature more detailed and accurate information. Consequently, the goal is to furnish websites with more comprehensive and precise information, thereby improving the quality and reliability of online health resources.

Supplemental material

Article information Continued