Content, quality, and reliability of spinal tuberculosis-related videos on Bilibili, TikTok, and Xiaohongshu: a cross-sectional analysis
This cross-sectional analysis of 214 spinal tuberculosis videos on Bilibili, TikTok, and Xiaohongshu reveals that while professionally produced content is generally of moderate quality and understandable, it often lacks actionable guidance, and popularity metrics fail to reliably predict information quality, highlighting the need for enhanced professional participation and platform-specific safeguards.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Spinal tuberculosis is a serious infection that attacks the bones of the spine. Unlike the more common form of the disease that affects the lungs, this version can silently destroy the vertebrae, leading to severe deformity, chronic pain, and even paralysis if not treated correctly. The cure exists, but it is a long and difficult road requiring strict adherence to a combination of powerful medicines for many months. If a patient stops taking the medication too soon or misunderstands the severity of their condition, the consequences can be permanent. In the modern world, people often turn to social media for answers about their health, hoping to find clear explanations and guidance. However, the internet is a vast ocean of information where expert advice sits alongside personal stories and unverified claims, making it difficult for a sick person to know which video to trust.
A team of researchers set out to map this digital landscape, specifically looking at how spinal tuberculosis is presented on three of China's most popular short-video platforms: Bilibili, TikTok (known there as Douyin), and Xiaohongshu. They wanted to see if the videos people were watching were actually helpful, reliable, and easy to understand. To do this, they gathered the first hundred videos returned by a search for "spinal tuberculosis" on each site, resulting in a collection of 300 clips. After removing duplicates, ads, and videos that did not focus on the disease, they analyzed 214 remaining videos. The researchers did not just watch them; they evaluated them using a set of standard tools designed to measure how good health information is. These tools checked for overall educational value, the reliability of the source, whether the video cited its references, how easy the language was to follow, and most importantly, whether the video told the viewer exactly what steps to take next.
The results revealed a mixed picture. The majority of the videos, about 81 percent, were created by professional doctors, and most of them were designed as short educational lessons rather than personal stories. The overall quality of the information was moderate; the videos were generally understandable, and most people could follow the medical concepts being explained. However, a significant gap emerged between understanding the information and knowing what to do with it. While three-quarters of the videos were clear enough to be understood, only a tiny fraction, about 12 percent, provided clear, actionable instructions. In other words, many videos explained what spinal tuberculosis was, but very few told the viewer clearly that they needed to see a specialist, start a specific long-term treatment, or watch for specific warning signs that required immediate help.
The platform where the video was found mattered less than who made it and how it was presented. Videos created by independent users without medical training consistently scored lower on quality and reliability than those made by doctors or medical organizations. The format of the video also played a role; structured educational lectures tended to be more reliable and easier to understand, while videos based on personal experience often lacked the necessary depth and accuracy. Interestingly, the researchers found that popularity was not a good indicator of quality. The videos with the most likes, comments, and shares were not necessarily the ones with the best medical information. In fact, the most popular platform, TikTok, generated the highest engagement, but its videos were not significantly better or worse in quality than those on the other sites. The number of views a video received seemed to reflect how entertaining or algorithmically favored it was, rather than how scientifically sound it was.
The study also highlighted a difference in the length and style of content across the platforms. Videos on Bilibili tended to be much longer, often resembling traditional lectures, while those on TikTok and Xiaohongshu were shorter and more concise. Despite these differences, the core finding remained consistent across all three sites: the information available was generally safe and understandable, but it often failed to give the viewer a clear path forward. The researchers concluded that while doctors and medical organizations are doing a good job of creating content, there is a need for these creators to be more explicit about the next steps a patient should take. They also suggested that the platforms themselves could do more to ensure that high-quality, medically accurate information is prioritized over content that is merely popular. For a patient searching for help with a serious condition, the goal should not just be to find a video that is easy to watch, but one that provides a clear, trustworthy guide to getting better.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.