Online information on chronic pain in 3 countries: an assessment of readability, credibility, and accuracy

Readability, reliability and quality of responses generated by ChatGPT, gemini, and perplexity for the most frequently asked questions about pain

It is clear that artificial intelligence-based chatbots will be popular applications in the field of healthcare in the near future. It is known that more than 30% of the world's population suffers from chronic pain and individuals try to access the health information they need through online platforms before applying to the hospital. This study aimed to examine the readability, reliability and quality of the responses given by 3 different artificial intelligence chatbots (ChatGPT, Gemini and Perplexity) to frequently asked questions about pain. In this study, the 25 most frequently used keywords related to pain were determined using Google Trend and asked to every 3 artificial intelligence chatbots. The readability of the response texts was determined by Flesch Reading Ease Score (FRES), Simple Measure of Gobbledygook, Gunning Fog and Flesch-Kincaid Grade Level readability scoring. Reliability assessment was determined by the Journal of American Medical Association (JAMA), DISCERN scales. Global Quality Score and Ensuring Quality Information for Patients (EQIP) score were used in quality assessment. As a result of Google Trend search, the first 3 keywords were determined as "back pain," "stomach pain," and "chest pain." The readability of the answers given by all 3 artificial intelligence applications was determined to be higher than the recommended 6th grade readability level (P < .001). In the readability evaluation, the order from easy to difficult was determined as Google Gemini, ChatGPT and Perplexity. Higher GQS scores (P = .008) were detected in Gemini compared to other chatbots. Perplexity had higher JAMA, DISCERN and EQIP scores compared to other chatbots, respectively (P < .001, P < .001, P < .05). It has been determined that the answers given by ChatGPT, Gemini, and Perplexity to pain-related questions are difficult to read and their reliability and quality are low. It can be stated that these artificial intelligence chatbots cannot replace a comprehensive medical consultation. In artificial intelligence applications, it may be recommended to facilitate the readability of text content, create texts containing reliable references, and control them by a supervisory expert team.

Unhelpful Information About Low Back and Neck Pain on Physiotherapist's Websites

BACKGROUND

The internet is increasingly used as a primary source of information for patients with musculoskeletal pain. Private physiotherapy practices provide informative content on low back pain (LBP) and neck pain (NP) on their websites, but the extent to which this information is biopsychosocial, guidelines-consistent, and fear-inducing is unknown. The aim of this study was to analyse the information on websites of private physiotherapy practices in the Netherlands about LBP and NP regarding consistency with the guidelines and the biopsychosocial model and to explore the use of fear-inducing language.

METHODS

The content of all existing Dutch private physiotherapy practice websites was examined in a cross sectional study design. Content analysis was based on predetermined criteria of the biopsychosocial model and evidence-based guidelines. Descriptive statistics were applied.

RESULTS

After removing duplicates and sites without information, 834 (10%) of 8707 websites remained. Information about LBP was found on 449 (54%) websites and 295 (35%) websites informed about NP. A majority of websites (LBP: n = 287, 64%; NP: n = 174, 59%) were biomedically oriented. Treatment advice was given 1855 times on n = 560 (67%) websites. Most of the recommended interventions were inconsistent with or not mentioned in the guidelines. Fear-inducing language was provided n = 1624 (69%) times.

CONCLUSIONS

The majority of the information on the Dutch private physiotherapy websites was biomedical and not in line with the current guidelines and fear-inducing.

SIGNIFICANCE

This study revealed that the majority of private physiotherapy practice websites that provide information on LBP and NP are inconsistent with the biopsychosocial model and current guidelines. This information has a negative impact on patients' knowledge, attitudes, beliefs, and expectations, potentially affecting treatment outcomes. Consequently, most physiotherapy practice websites are unreliable sources of information on LBP and NP for patients. There is an urgent need to implement strategies to make websites reliable and helpful sources of information.

Credibility, readability and content analysis of treatment recommendations for adolescents with nonspecific back pain published on consumer websites

BACKGROUND

Parents may seek out health information online when their adolescent has nonspecific back pain to better understand treatment options. Such information directed towards consumers has not been previously analysed.

METHODS

A descriptive cross-sectional content analysis was performed to describe the treatments recommended on consumer websites for nonspecific back pain in adolescents. The credibility and readability of the websites were also assessed. Systematic Google searches were completed in five countries, and relevant content from eligible web pages was analysed. An a priori codebook with 34 treatment-related codes was developed. Nine additional codes were inductively created during analysis. Credibility was assessed using the JAMA benchmark. Readability was assessed via the Flesch Kincaid Grade Level.

RESULTS

Of 245 web pages, 48 were deemed eligible and analysed. Of 43 treatment codes, 37 were present in at least one web page. The five most frequently identified codes were See the doctor/get a diagnosis (found on 85% of web pages), Ergonomics/posture/biomechanics (52%), Reassurance (48%), Physiotherapy (48%) and Non-prescription pharmaceuticals/supplements (46%). Only 21% of the web pages met all four JAMA benchmark criteria, and 15% cited at least one recent or high-quality source. The median Flesch Kincaid Grade Level score was 9.0 (range 3.5-12.9).

CONCLUSIONS

Parents of adolescents with nonspecific back pain may find that treatment recommendations published online are numerous and varied, with visits to the doctor encouraged. The credibility scores of these web pages are generally low, while the median reading level may be too high for the general population.

SIGNIFICANCE STATEMENT

This analysis reveals that public-facing websites with recommendations for treating adolescent nonspecific back pain do not cite the most recent, high-quality research. Although web pages correctly encourage physical activity and exercise over surgery and prescription medications, they do not reflect the psychologically informed or interdisciplinary care emphasized in recently published treatment recommendations. Clinicians must be aware that caregivers of their adolescent patients with nonspecific back pain may be exposed to online messages that encourage them to keep seeking a diagnosis.

How Low Back Pain is Managed-A Mixed-Methods Study in 32 Countries. Part 2 of Low Back Pain in Low- and Middle-Income Countries Series

BACKGROUND: The Lancet Low Back Pain (LBP) Series highlighted the lack of LBP data from low- and middle-income countries (LMICs). The study aimed to describe (1) what LBP care is currently delivered in LMICs and (2) how that care is delivered. DESIGN: An online mixed-methods study. METHODS: A Consortium for LBP in LMICs (n = 65) was developed with an expert panel of leading LBP researchers (>2 publications on LBP) and multidisciplinary clinicians and patient partners with 5 years of clinical/lived LBP experience in LMICs. Quantitative data were analyzed using descriptive statistics. Two researchers independently analyzed qualitative data using inductive and deductive coding and developed a thematic framework. RESULTS: Forty-seven (85%) of 55 invited panel members representing 32 LMICs completed the survey (38% women, 62% men). The panel included clinicians (34%), researchers (28%), educators (6%), and people with lived experience (4%). Pharmacotherapies and electrophysiological agents were the most used LBP treatments. The thematic framework comprised 8 themes: (1) self-management is ubiquitous, (2) medicines are the cornerstone, (3) traditional therapies have a place, (4) society plays an important role, (5) imaging use is very common, (6) reliance on passive approaches, (7) social determinants influence LBP care pathway, and (8) health systems are ill-prepared to address LBP burden. CONCLUSION: LBP care in LMICs did not consistently align with the best available evidence. Findings will help research prioritization in LMICs and guide global LBP clinical guidelines. J Orthop Sports Phys Ther 2024;54(8):560-572. Epub 11 April 2024. doi:10.2519/jospt.2024.12406.

It's OK to Move! Effect of a Brief Video on Community Confidence in Activity Despite Back Pain: A Randomized Trial

OBJECTIVE: To estimate the difference in confidence to become active despite low back pain in people who were exposed to one of 2 video interventions delivered on social media, compared to no intervention. DESIGN: A proof-of-concept, 3-group randomized controlled trial, in a 1:1:1 ratio. METHODS: Participants aged 18 years and over, with and without low back pain, were recruited via the social media channel Facebook, to view either a humorous video, a neutral video, or to no intervention. The videos were delivered online, explained evidence-based management for low back pain, and were designed to "go viral." The primary outcome was confidence in becoming active despite pain, measured using the Pain Self Efficacy Questionnaire (Item 10) (ranges from 0 [not at all confident] to 6 [completely confident]) immediately after watching the video. We aimed to capture the real-time impact and immediate reactions that contributed to the content's reach. RESULTS: Among 1933 randomized participants (mean [standard deviation] age: 58.9 [14.0] years, 1285 [75%] women), 1232 [70%] had low back pain and 88.8% completed the primary outcome. One thousand two hundred sixty-four participants were randomized to receive a video intervention, and 633 participants did not receive a video. On a 6-point scale, individuals exposed to either video (n = 1088) showed a mean confidence level 0.3 points higher (95% confidence interval: 0.1, 0.6) compared with no video (n = 630). CONCLUSION: Participants who viewed a brief video intervention reported a very small difference in confidence to become active despite low back pain, compared with no intervention. The difference may lack clinical relevance. J Orthop Sports Phys Ther 2024;54(6):1-8. Epub 18 April 2024. doi:10.2519/jospt.2024.12412.