Platelet-rich plasma videos for knee osteoarthritis on YouTube are of acceptable quality and content

Abstract

Individuals seeking information about platelet-rich plasma (PRP) treatment for knee osteoarthritis, a controversial topic, may turn to YouTube. However, the quality and content adequacy of these videos should be questioned. This study analyzed YouTube videos related to PRP to evaluate their quality and content. The primary aim was to assess these videos’ quality and content, while the secondary aim examined the relationship between video quality/content and video characteristics, information sources, and themes. A total of 250 videos related to PRP treatment for knee osteoarthritis were identified through YouTube searches, and 68 of them were included in the study. The number of views, number of days since upload, view ratio, video duration, number of comments, the source of the information presented in the video, and the overall theme of the video were recorded. The quality and content of the videos were evaluated using the DISCERN, Journal of the American Medical Association, Global Quality Score, and PRP scoring systems. Median scores for video quality and adequacy were: DISCERN: 48 (range: 19–76), Journal of the American Medical Association: 2 (range: 1–4), Global Quality Score: 3.5 (range: 1–5), and PRP score: 6 (range: 2–10). Video quality and content positively correlated with the number of views, view ratio, duration, and comments. Videos in which patients provided the information had the lowest scores in terms of quality and content. While sports medicine physicians generally achieved higher scores, no difference was found between orthopedic surgeons and other specialized physicians. This study has shown that YouTube videos related to PRP treatment for knee osteoarthritis are not of low quality. The positive correlation between video characteristics, such as the number of views and comments, and quality suggests that the YouTube algorithm may prioritize higher-quality content, potentially enabling individuals to access better videos. Physicians provided higher-quality content compared with patients, although differences were observed among medical specialties. This finding suggests that the quality of informative content on PRP may vary depending on the expertise of the content creator. Despite certain limitations, YouTube may serve as a useful resource for individuals seeking information about PRP treatment. However, for the most reliable information, academic sources with an editorial process should be consulted.

1. Introduction

With the advancement of technology, the ease of accessing information has made the internet an important tool for addressing patients’ curiosity and concerns about health. One of the most significant video-based platforms for accessing information on the internet is YouTube. Although YouTube primary purpose is not health-related education, it has been observed that health-related content has increased over time due to the interest of both patients and doctors.^[1] Despite the prevalence of health-related videos on YouTube, the platform does not provide an editorial process to evaluate the quality and content of this information. Therefore, there may be concerns regarding the reliability of the information obtained from this source.

Knee osteoarthritis is a pathology that affects a large portion of society. Despite significant advancements achieved through surgical treatments, there is still no consensus on a gold standard treatment for conservative management. The ambiguity regarding treatment options drives both doctors and patients to seek alternative treatments. One such treatment is platelet-rich plasma (PRP) therapy.^[2] When doctors recommend PRP therapy to patients with knee osteoarthritis, patients may seek additional information about this treatment. Shared decision-making is crucial in determining treatments in the healthcare field, and with the help of the internet, patients can quickly access different perspectives for obtaining a second opinion. One of the sources they can use to access information online is YouTube.^[3] Although numerous studies in the literature have assessed the quality and content of information obtained from YouTube regarding orthopedic diseases and treatment methods,^[1,4–17] no study has been conducted specifically on PRP therapy for knee osteoarthritis, which is a highly relevant topic.

In this study, videos on PRP therapy for knee osteoarthritis available on YouTube were examined and analyzed. To the best of our knowledge, this is the first study in the literature on this topic. The primary aim of this study is to evaluate the overall quality and adequacy of the content of YouTube videos. The secondary aim is to assess whether the quality and adequacy of the videos are associated with video characteristics, the sources providing the information, and the themes of the video content.

2. Methods

This study was conducted based on PRP videos related to knee osteoarthritis on YouTube, and institutional review board approval was not required for this analysis of publicly available data. On January 29, 2025, a search was conducted on YouTube’s search engine using the keywords “Platelet rich plasma knee,” “Platelet rich plasma knee pain,” “Platelet rich plasma knee injection,” “Platelet rich plasma knee osteoarthritis,” and “Platelet rich plasma gonarthrosis,” sorted by relevance. To prevent YouTube’s search algorithm from influencing the results, Incognito mode on the Google Chrome browser was used. Incognito mode ensures that watch history, location, and demographics do not affect the search query outcomes.^[14] After each search, the first 50 videos with English titles were selected. The decision to use the first 50 videos was based on previous similar YouTube-based studies that also analyzed 50 videos.^{[14,15,17–19]} During the selection process, “Youtube Shorts” videos and videos shorter than 60 seconds were excluded due to the likelihood of providing insufficient information, and the next eligible videos were selected instead. Among the 250 initially selected videos, 149 were excluded due to duplication, and 6 were excluded for not having English audio. From the remaining 95 videos, 27 were excluded as they were not related to knee osteoarthritis. In total, 68 videos were analyzed (Fig. 1).

Figure 1.

Flowchart illustrating the inclusion and exclusion criteria for YouTube videos analyzed in the study.

For each video, the title, number of views, number of days since upload (as of January 29, 2025), view ratio (number of views/number of days since upload), video duration (in seconds), and the number of comments were recorded. The presenters and the topics discussed in the videos were also documented. Based on the presenters’ qualifications, the video sources were classified into 4 categories: orthopedic surgeons, sports medicine physicians, other specialized physicians, and patients. The video themes were categorized into 3 groups: general information, injection techniques, and patient testimony.

For the quality and content evaluation of the videos, DISCERN, the Journal of the American Medical Association (JAMA) Benchmark, the Global Quality Score (GQS), and a “PRP scoring system” developed by the researchers were used. These assessments were conducted independently by 2 orthopedic surgeon authors, and interobserver reliability was evaluated using the intraclass correlation coefficient (ICC). The average scores from both authors were used for further analysis. The DISCERN scoring system consists of 16 questions, with each question rated on a scale from 1 to 5. The minimum possible score is 16, and the maximum is 80, with higher scores indicating higher quality. The first 8 questions assess reliability, the next 7 focus on the specific details of the treatment, and the final question evaluates overall quality. The JAMA benchmark score consists of 4 criteria: authorship, attribution, currency, and disclosure, with each criterion scored between 0 and 1. The highest possible score is 4, indicating a highly credible source. The GQS is a 5-point rating system used to assess the overall educational value of the content, with a score of 5 representing high-quality content. The PRP scoring system, developed by the study authors based on previous studies,^{[4,6–8,10,11,13,20]} consists of 10 questions designed to assess general knowledge about PRP. The possible scores range from 0 to 10 (Table 1).

Table 1.

PRP score.

Questions	1: Information provided 0: No information
What is PRP?
How is it obtained?
Who is it applied to?
Who is not it applied to?
What is it used for? What is the goal of PRP treatment?
How does it work/what is the mechanism of action?
How is it applied?
What are the risks?
What should be considered after the procedure?
Are there other treatment options besides PRP?

PRP = platelet-rich plasma.

2.1. Statistics

For the assessment of DISCERN, JAMA, GQS, and PRP scores ICC (model: two-way fixed, type: absolute agreement) was used to analyze the 2 reviewers’ agreement on each scoring system. The 2 reviewers’ scores were averaged and used for further analysis. Correlation was classified as poor (0.00–0.20), fair (0.21–0.40), moderate (0.41–0.60), good (0.61–0.80), or excellent (0.81–1.00).^[4,20] Assumptions of normality were checked by using Shapiro–Wilk and Levene tests, nonparametric tests were utilized since these assumptions were not fulfilled. After scoring each video, the correlations among the scores and between video characteristics and scores were analyzed using the Spearman ρ test. Differences in scores based on the video source and video theme were examined with the Kruskal–Wallis test, and in cases of statistically significant results, post hoc comparisons were carried out with adjusted P values. The significance levels were set at P < .05 for all tests. All analyses were carried out by using SPSS 22.

3. Results

The median view count of the 68 videos was 6656.5 (range: 12–764,150), and the median number of days since upload (as of January 29, 2025) was 900 (range: 32–4706) days. The median view ratio was 5.27 (range: 0.01–2247.50), the median duration was 294 seconds (range: 60–4143), and the median number of comments was 19.5 (range: 0–1539). Since comments were disabled for 4 videos, comment analyses were conducted on 64 videos. ICC for interobserver reliability was found to be 0.88 for DISCERN, 0.83 for JAMA, 0.88 for GQS, and 0.94 for PRP scores, indicating excellent reliability. The median scores for evaluating the quality and adequacy of the videos were as follows: DISCERN: 48 (range: 19–76), JAMA: 2 (range: 1–4), GQS: 3.5 (range: 1–5), and PRP Score: 6 (range: 2–10) (Table 2). A positive correlation was observed among DISCERN, JAMA, GQS, and PRP scores (P < .001). In addition, view count, view ratio, duration, and number of comments showed a positive correlation with these scores (P < .05) (Table 3).

Table 2.

Overall video characteristics (view count, number of days since upload, view ratio, duration, comments) and quality-content scores (DISCERN, JAMA, GQS, PRP score).

	Median (range)		Median (range)
View count	6656.5 (12–764,150)	DISCERN	48 (19–76)
Days (number of days since upload)	900 (32–4706)	JAMA	2 (1–4)
View ratio	5.27 (0.01–2247.50)	GQS	3.5 (1–5)
Duration (second)	294 (60–4143)	PRP score	6 (2–10)
Comments*	19.5 (0–1539)

GQS = Global Quality Score, JAMA = Journal of the American Medical Association, PRP = platelet-rich plasma.

^*Comments were analyzed based on 64 videos, as comments were disabled for 4 videos.

Table 3.

Correlation of video characteristics and quality-content scores.

	DISCERN (P, ρ)	JAMA (P, ρ)	GQS (P, ρ)	PRP score (P, ρ)
View count	.005, 0.339	.003, 0.350	.006, 0.327	.022, 0.278
View ratio	<.001, 0.443	<.001, 0.439	<.001, 0.418	.009, 0.315
Duration (seconds)	<.001, 0.756	<.001, 0.546	<.001, 0.719	<.001, 0.604
Comments	<.001, 0.517	<.001, 0.498	<.001, 0.503	.008, 0.331

rho: Spearman ρ value. Statistical significance at the P < .05 level.

GQS = Global Quality Score, JAMA = Journal of the American Medical Association, PRP = platelet-rich plasma.

Regarding the video source, 19 videos were presented by orthopedic surgeons, 26 by sports medicine physicians, 13 by other specialized physicians, and 10 by patients. In terms of the themes of the videos, 54 provided general information, 10 featured patient testimony, and 4 focused on injection techniques. Kruskal–Wallis analysis revealed significant differences in DISCERN (P < .001), JAMA (P < .001), GQS (P < .001), and PRP scores (P = .001) based on the video source (Table 4). In the post hoc analysis, sports medicine physicians scored higher than orthopedic surgeons in DISCERN (P = .035), but no difference was found between sports medicine physicians and other specialized physicians (P = .183). There was no difference in DISCERN scores between orthopedic surgeons and other specialized physicians (P > .99). For JAMA scores, sports medicine physicians scored higher than both orthopedic surgeons (P = .001) and other specialized physicians (P = .035). No difference in JAMA scores was found between orthopedic surgeons and other specialized physicians (P > .99). For GQS and PRP scores, no significant differences were found among sports medicine physicians, orthopedic surgeons, and other specialized physicians (P > .05). Patients had the lowest overall scores in DISCERN, JAMA, GQS, and PRP (P < .05).

Table 4.

Comparison of scores by video source (orthopedic surgeons, sports medicine physicians, other physicians, and patients).

	Orthopaedic surgeons (n = 19)	Sports medicine physicians (n = 26)	Other physicians (n = 13)	Patients (n = 10)	P value*
DISCERN	44.0 (25.5–63.0)	62.5 (28.0–76.0)	47.0 (35.0–69.0)	36.0 (19.0–39.5)	P < .001
JAMA	2.0 (1.5–4.0)	3.5 (2.0–3.5)	2.0 (2.0–3.0)	1.5 (1.0–2.0)	P < .001
GQS	3.5 (1.5–4.5)	4.5 (2.0–5.0)	3.5 (2.0–4.5)	2.0 (1.0–2.5)	P < .001
PRP score	6.0 (4.5–8.5)	6.5 (3.0–10.0)	6.5 (4.5–9.0)	3.25 (2.0–7.0)	P = .001

Data presented as median (range).

GQS = Global Quality Score, JAMA = Journal of the American Medical Association, PRP = platelet-rich plasma.

^*Kruskal–Wallis test.

Kruskal–Wallis analysis showed significant differences in DISCERN, JAMA, GQS, and PRP scores based on the video themes (P < .001). Testimony videos scored lower than general information videos across all scoring systems (P < .001). No differences were found between injection and technique videos and testimony videos in any of the scoring systems (P > .05). Compared with general information videos, injection and technique videos had lower DISCERN (P = .009) and GQS (P = .014) scores. However, injection and technique videos did not differ from general information videos in JAMA (P = .505) and PRP (P = .179) scores.

4. Discussion

With the rapid development of technology today, access to information has become easier than ever. The widespread use of digital devices, such as smartphones, tablets, and computers, has led individuals to prefer the internet, which is one of the primary sources they turn to for health-related information. Especially, social media platforms and video-sharing websites enable users to quickly and visually access health-related content. While YouTube has become a popular resource for patients seeking health information, the reliability and quality of the content on this platform remain controversial. Knee osteoarthritis is a condition that affects a large portion of the population worldwide and can lead to significant functional loss. The wide range of treatments, from lifestyle changes to surgical interventions, may lead both doctors and patients to search for various treatment options. One such treatment is PRP therapy.^[21] Patients experiencing knee pain may want to research PRP before consulting a doctor or after receiving their doctor’s recommendation. One of the important sources individuals turn to for information on this topic is video content on YouTube. This suggests that the quality and content of videos related to PRP in knee osteoarthritis should be investigated, as low-quality information obtained from YouTube can have negative effects on the doctor–patient relationship.^[22] Therefore, in this study, YouTube videos on PRP application in the treatment of knee osteoarthritis were examined and analyzed. The primary aim of the study is to assess the overall adequacy of these videos in terms of quality and content. In addition, the second aim is to examine the relationship between video quality and content scores and various video characteristics (such as the number of views, view ratio, video duration, and number of comments), sources of information, and video themes. To the best of our knowledge, there are no other studies on this topic in the literature, and this study is the first to address this issue.

In this study, the findings indicate that YouTube videos related to PRP treatment for knee osteoarthritis are not of poor quality in terms of both content and overall quality. In addition, it was observed that among these videos, some had very high scores in both quality and content and had also reached a high number of views. Several studies in the literature have investigated the reliability of YouTube as an information source on orthopedic conditions and their treatments. Some of the research topics include videos on bankart lesions,^[1] rotator cuff tears,^[4,11,12] superior labrum anterior and posterior (SLAP) lesions,^[14] reverse shoulder arthroplasty,^[8] anterior cruciate ligament tears,^[6] kyphosis,^[7] disc herniation,^[9] femoroacetabular impingement,^[13] medial patellofemoral ligament,^[16] hip arthroscopy,^[10] carpal tunnel syndrome,^[15] and hallux valgus.^[17] A consistent conclusion across these studies is that the information presented on YouTube is generally inadequate and of low quality. In addition, some studies attribute this low quality to the platform’s lack of an editorial review process, which allows anyone to upload content freely.^[8,11] However, this study found that videos related to PRP application in knee osteoarthritis are not of low quality. Therefore, YouTube may provide information that cannot be considered poor in terms of patient education strategies regarding PRP and may help clarify many aspects of the topic. However, it should not be overlooked that the median scores were not at an excellent level and that the information obtained from YouTube may still be insufficient. Thus, we believe that, in clinical practice, patients should consult their own physicians, affiliated healthcare institutions, and scientific sources to support informed decision-making.

One significant finding of this study is the observed positive correlation between video quality and content scores and various video metrics, including view count, view ratio, video duration, and number of comments. In this context, our findings show some significant differences compared with previous studies. For instance, Celik et al^[4] in their study evaluating YouTube videos on rotator cuff repair, found that viewers tended to favor videos with lower scores, poorer quality, and shorter durations. The preference of viewers for low-quality videos is not confined to orthopedic conditions; similar trends have been observed in studies on other medical topics as well.^[23] MacLeod et al,^[13] in their study on femoroacetabular impingement, and Mert and Bozgeyik,^[15] in their analysis of carpal tunnel syndrome videos, both concluded that video quality scores did not correlate with view counts or other video characteristics. Likewise, Matzko et al,^[14] examining videos on SLAP lesions, expressed concern over the lack of association between quality scores and viewer engagement, noting that high-quality videos were not viewed more frequently than those of lower quality. However, in our study, video characteristics showed a positive correlation with quality and content scores. These findings allow us to make 3 key inferences. First, individuals seeking information about PRP treatment for osteoarthritis tend to engage with higher-quality videos in terms of both content and overall quality. Second, longer videos tend to be superior in both content and quality. Third, this finding highlights an important point: YouTube operates based on algorithms. YouTube’s ranking algorithm evaluates multiple factors to highlight the most relevant content, with view count being a significant determinant. Therefore, when a viewer searches for information about PRP application in knee osteoarthritis, the first videos they encounter are likely to have higher view counts and, consequently, a higher probability of being of better quality.

In this study, the information sources in the videos, that is, the individuals providing the information, were also analyzed, and controversial results were obtained. Previous studies, including one by Celik et al,^[4] have pointed out that, although the information provided by physicians is of higher quality compared with other sources, it is still considered insufficient in terms of quality. In Matzko et al’s^[14] study on SLAP lesions and Jildeh et al’s^[10] study on hip arthroscopy, it was similarly noted that the videos provided by physicians were of higher quality. However, there are also studies that have obtained different results. In Fares et al’s^[8] study on reverse shoulder arthroplasty, it was found that academic and physician sources did not provide videos of superior quality compared with other sources, and they noted that this was surprising. Similarly, in Mert and Bozgeyik^[15] study on carpal tunnel syndrome and Tekin and Bozgeyik^[17] study on hallux valgus, it was found that the videos provided by physicians and academic sources were not of superior quality compared with other sources. These previous studies show us that whether the videos prepared by physicians are of higher quality compared with other sources may vary depending on the orthopedic disease and topic. However, our study draws attention to a completely different point. In our study, the individuals providing the information in the videos were categorized into 4 groups: sports medicine physicians, orthopedic surgeons, other specialized physicians, and patients. One of the key differences from previous studies is that we categorized physicians based on their specialties rather than as a whole group. According to the study results, physicians generally provide better information than patients, but when comparing physicians within their own group, sports medicine physicians stood out in some of the scoring categories. While the exact reason for this is difficult to determine, some hypotheses can be made. PRP application for knee osteoarthritis is a topic that attracts the interest of various medical specialties. Orthopedic surgeons, sports medicine physicians, and physical medicine and rehabilitation specialists are the main specialties that apply this treatment. While orthopedic surgeons may approach knee osteoarthritis from a more surgical-centered perspective, sports medicine physicians may be more focused on conservative treatments. Therefore, it is likely that sports medicine physicians can produce higher-quality content regarding PRP treatments for knee osteoarthritis. Thoughts that surgeons, as the group performing surgical treatments, may show deficiencies in producing informative videos have also been raised in other studies.^[17,20]

This study has several important limitations. The first is that, like in many other social media-based studies, the dynamic nature of YouTube means that content can change over time. Videos may be deleted or new ones may be added. Over time, the quality related to PRP treatment and knee osteoarthritis could change based on newly uploaded videos. Another limitation is that even though we used 5 different keywords for the place of PRP in knee arthritis treatment, patients may conduct different searches from those we used, potentially leading them to different videos. Therefore, even though an evaluation was made based on 250 videos, a complete representation may not be achieved. Another limitation of this study is the exclusion of short-form videos. Although such content has become increasingly popular among users due to its accessibility and brevity, it was excluded based on the assumption that the limited duration would not allow for the presentation of sufficient medical information or for meaningful assessment using scoring systems such as DISCERN, JAMA, GQS, and the PRP score. Future studies may consider analyzing short-form video content separately to determine its potential educational value and its impact on viewer understanding in the context of medical information. Another limitation is that the scoring was performed by only 2 evaluators. The fact that the scoring in this study was done by only 2 authors may be considered insufficient. However, previous studies have also conducted evaluations based on scores given by 1^[11] or 2^{[1,13–15,20]} evaluators. In addition, the study showed excellent agreement between evaluators in terms of ICC. Furthermore, the ICC analysis conducted in this study, unlike other studies, was done using “absolute agreement” rather than “consistency”^[1,14] and the use of the average score in the score analysis is a factor that improves the quality of the evaluation. Another important limitation is that YouTube no longer displays the like/dislike count, so it is not possible to provide clear information about whether viewers liked the videos or not. Therefore, it is impossible to determine how much the videos that received high scores satisfied the viewers. Another limitation concerns the PRP score. This scoring system is a nonvalidated assessment tool that we developed to evaluate content related to PRP. One of the main reasons for using this scoring system was to complement the DISCERN, JAMA, and GQS scores with a tool that could provide a more specific evaluation of the content of PRP-related videos. While this scoring system assesses important points related to PRP, it may be inadequate. Nevertheless, it is worth noting that similar nonvalidated assessment tools developed by authors have been used in previous studies and are accepted in the literature for evaluating orthopedic video content.^{[4,6–8,10,11,13,20]}

To conclude, this study has shown that YouTube videos related to PRP treatment for knee osteoarthritis are not of low quality or content. The findings suggest that YouTube can provide individuals researching PRP with information of moderate-to-good quality. In particular, the positive correlation between video characteristics such as the number of views and comments, and the quality and content scores, suggests that YouTube’s algorithm may prioritize higher-quality content for individuals searching for this topic. In addition, analysis of the sources of information in the videos revealed that physicians provide higher-quality content compared with patients, but differences based on the physician’s area of expertise were also observed. Sports medicine physicians were found to present higher-quality content in videos related to PRP treatment for knee osteoarthritis. This suggests that the informative content about PRP treatment may vary depending on the specialty due to differences in the approach to treatment. Although the study has some limitations, the results indicate that YouTube can be a valuable source of information for individuals seeking information about PRP treatment. However, it should be noted that for the most accurate and reliable information, individuals should consult their doctors, academic sources, and trusted healthcare organizations. Future studies could provide valuable contributions to the more comprehensive evaluation of health content on YouTube and the development of verification mechanisms.

Acknowledgments

We would like to thank Prof. Coskun Kus PhD for their valuable guidance on the statistical analyses.

Author contributions

Conceptualization: Ebubekir Eravsar.

Data curation: Ebubekir Eravsar, Numan Mercan.

Formal analysis: Ebubekir Eravsar.

Investigation: Ebubekir Eravsar, Numan Mercan.

Methodology: Ebubekir Eravsar.

Project administration: Ebubekir Eravsar.

Visualization: Ebubekir Eravsar.

Writing – original draft: Ebubekir Eravsar, Numan Mercan.

Writing – review & editing: Ebubekir Eravsar, Numan Mercan.