“My knee is cracking” – What information is available on the internet about it? A systematic appraisal of the credibility, readability and accuracy of online information about knee crepitus

Abstract

Objectives

To summarise and systematically appraise the credibility, readability, and accuracy of the online information about knee crepitus.

Design

Systematic appraisal of online information.

Method

The top 20 URLs of two search engines (Google and Bing) for the following terms were screened for eligibility ‘knee crepitus’, ‘knee sound’, and ‘knee crackling’ (N ​= ​120 websites). Two reviewers assessed the websites for credibility (JAMA benchmark) and readability (Flesch Reading Ease and Flesch-Kincaid Grade Level tests). For accuracy, we performed a qualitative analysis creating thematic units based on the website's content and explored whether they were supported by evidence.

Results

Fifty-one websites were included. There was a large variation in the overall credibility of websites, with more than half of the websites lacking credible information. The median Flesch reading ease was 62 (range ​= ​42–88) and Flesch–Kincaid grade level was 8 (range ​= ​3–12). Six key thematic units emerged from our qualitative analysis all with sub-themes within them. The main topics covered by websites were: (i) What is knee crepitus? (ii) Is knee crepitus a problem? (iii) What is the cause of knee crepitus? (iv) How to manage knee crepitus? (v) What are the health consequences of having knee crepitus? (vi) Who should I seek if I have knee crepitus?

Conclusion

Patients and clinicians should take caution when seeking web-based information about knee crepitus. Websites about knee crepitus lack credibility, present a large variation in readability outcomes, and there was no research evidence available to support most of their information.

1. Introduction

The Internet is one of the most common sources for the general public to obtain health information [1]. The number of web searches seeking information related to signs and symptoms of musculoskeletal conditions has increased more than 5 times in the past decade [2]. The Internet offers convenient and rapid access to health information; however, misleading content can negatively impact the decision-making and behaviour of people with musculoskeletal conditions (e.g. online information promoting rest for people with low back pain [3], the use of ice for patellofemoral pain [4,5], or opioids for knee osteoarthritis [6]).

A common clinical feature across many musculoskeletal knee conditions is joint crepitus, including patellofemoral pain [4], meniscal impairments [7], and knee osteoarthritis [8]. Crepitus is often described by patients as ‘grinding’, ‘cracking’, and ‘clunking’ associated with knee flexion-extension movements, and figures among the primary complaints from patients with knee conditions [9]. Knee crepitus is associated with impaired quality of life [8] and future development of symptomatic knee osteoarthritis [10]. However, in people with patellofemoral pain, knee crepitus is not related to function, physical activity, pain, or knee strength [11]. Qualitative reports exploring patients' beliefs about their knee crepitus, reveal crepitus as a poorly understood symptom that creates negative emotions, inaccurate etiological beliefs and ultimately leads to altered behaviour [9,12,13]. Overall, participants reported poor experiences with health professionals, and despite them describing their crepitus as a major symptom, it was often disregarded. This often leads patients to source information about knee crepitus on the Internet [4,9].

The credibility, readability, and accuracy of online information about knee crepitus are yet to be investigated. These are important considering the large uptake of patients in seeking health information on the Internet [2,14]; and, that inaccurate biomedical beliefs about clinical conditions are known to impact patient behaviour and decision-making [15]. We aimed to summarise and systematically appraise the online information about knee crepitus and evaluate its credibility, readability, and accuracy.

2. Methods

This study is a cross-sectional systematic appraisal of information available on the internet. Reporting of our findings followed the Strengthening the Reporting of OBservational studies in Epidemiology (STROBE) [16]. Ethics approval was not required for this study.

Search strategy: Searches were performed using two search engines, Google and Bing. These search engines were selected due to their widespread use (estimated unique monthly visitors of 1.8 billion and 500 million, respectively) [4]. Three search terms used were: (i) ‘Knee crepitus’, (ii) ‘Knee sound’ and (iii) ‘Knee cracking’. ‘Knee crepitus’ was chosen as it is the appropriate clinical term described in clinical practice guidelines [17,18]. ‘Knee sound’ and ‘knee cracking’ were used as they are more simplistic terms that may be utilised by health information seekers, with a Google Trends™ analysis suggesting ‘knee sound’ and ‘knee cracking’ are searched approximately 62 ​% and 46 ​% more than ‘knee crepitus’, respectively. We did not include the term ‘noisy knees’ in our search as it is not a term commonly searched according to our Google Trends™ analysis (87 ​% less than ‘knee crepitus’).

One member of the research team (FS) performed the searches and extracted website URLs in June 2023, from Melbourne, Australia. Prior to performing the searches, browsing history, cookies and temporarily cached files were deleted, no user accounts were logged in to avoid user-automated preferences (i.e. Google account) and geographical location was turned off [4,6]. The first 20 URLs for each search term on each search engine were included for screening (N ​= ​120 websites). This is in line with previous web-content systematic appraisals [4,6,19] and is based upon an assumption that health information seekers are most likely to view websites that appear on the first page of search results.

Eligibility criteria and screening: For this systematic appraisal, we included websites with content about knee crepitus. The information about knee crepitus should be freely available on the websites for the public. Websites with links to other forms of downloadable content, such as booklets, leaflets, or brochures, were also included. We excluded URLs directing to video-sharing platforms (e.g. YouTube), social media posts (e.g. X and Facebook), and peer-reviewed publications as they are not considered a “website about knee crepitus”, plus many articles are behind paywalls that may not be accessible to most people searching for information. Websites that were duplicated, inactive, had content unrelated to knee crepitus, or where the content was behind a paywall were also excluded. In cases of websites where the content was partially behind a paywall, only the free content was assessed. Two independent reviewers (FS and DAG) screened the list with 120 URLs and applied the eligibility criteria (supplementary file 1). A third reviewer (MFP) was available to solve any disagreements.

2.1. Data extraction

Basic characteristics of included websites were extracted by two members of the research team (FS and DAG), including whether it was Australia-based (i.e.com.au or.org.au address) or clearly focussed on advertising a product or service (commercial websites). Websites were classified as commercial if there was a direct link to purchase a product or sell a service (e.g. ‘book an appointment’). For each website, the content that was specific to knee crepitus was extracted to a pre-formatted template.

2.2. Credibility

Credibility was assessed using the Journal of the American Medical Association (JAMA) benchmark [19]. The JAMA benchmark consists of 4 elements: (1) currency of information, (2) declaration of authorship, (3) presentation of a list of references (attribution), and (4) disclosure of any conflict of interest, funding, or sponsorship. Each item was categorized as ‘yes’ or ‘no’. Each website was assessed by two independent reviewers (FS and DAG), and a third reviewer was available to solve any disagreements (MFP). The website was up to date if its date of publication or last update were no older than 2020 (approximately 3 years from our search). Authorship was declared when single or multiple authors were listed or when authorship was attributed to a working group. References were categorized as ‘yes’ if they were from peer-reviewed scientific journals. Links to blogs and newspapers were not considered as references. [19].

2.3. Readability

Two measures of readability were used in this study, the Flesch reading ease score [20] and the Flesch– Kincaid grade level [21]. Both measures of readability were selected due to their common use in health information research [22,23] and are calculated based on the average number of syllables per word and the average number of words per sentence. For reading ease score, a score from 0 to 100 is generated, with higher scores indicating easier readability [20]. For grade level, a score is calculated that is associated with reading difficulty equivalent to US school system grade ages [21]. Current guidelines suggest that health information should be communicated at 7–8th grade level (suitable for ages 12–14 years with 8 years of education). We used the following online tool in June 2023 (https://www.webfx.com/tools/read-able/). to assess the readability of each website, in line with previous research [22].

2.4. Sensitivity analyses

We compared the credibility and readability of websites that were Australia-based (i.e.com.au or.org.au), against those that were not. Despite taking precautions to avoid geographical bias, the location from where we performed the searchers could have affected our outputs. We also compared the credibility and readability of websites that were advertising or selling a product/service (commercial websites) against those that were not.

2.5. Qualitative analysis of knee crepitus content

The text information for each URL was extracted to a text editor to complete the qualitative analysis. Qualitative data were first analysed inductively using reflexive thematic analysis to generate thematic units [24]. More specifically, one researcher (FS) read and reread each text related to knee crepitus extracted from the websites to gain familiarity and develop initial thematic units of meaning and a coding structure. A second researcher (DAG), also read, reread, and independently coded the extracted text. The researchers met numerous times for reflective discussion, along with two other senior members of the research team (MFP and DOS), with the goal of (i) developing a thematic framework by mapping the information reported by the websites and refining thematic units; (ii) exploring the data, which involved careful reading and organisation of the data into sub-themes; and (iii) data interpretation and summary. All members of the research team revised and approved the thematic units created during data analysis.

2.6. Accuracy

We explored whether each category from the thematic units raised by the qualitative analysis of the websites was supported by scientific evidence. Two independent reviewers retrieved all references cited by the websites and the full reference list of all the articles. Our research team have large experience in the knee crepitus field [8,[11], [12], [13],25] and other references familiar to the research team were also added to the list (supplementary file 2). The sub-themes from the thematic units were classified as ‘accurate’ when the information was appropriately supported by one or a group of articles; ‘inaccurate’ when there was one or a group of articles providing evidence contrary to the information; and ‘partially accurate’ when the information was partially supported by one or a group of articles. We classified a category as ‘unable to determine’ when there was no evidence to support the sub-theme.

2.7. Data presentation

We used descriptive statistics to summarise credibility and readability across the websites and calculated the agreement rate (%) between reviewers for the screening and credibility assessment before final agreements were made. The qualitative thematic units are reported narratively followed by supporting quotes, and whether each category from the thematic units was accurate, inaccurate, partially accurate or unable to determine.

3. Results

Of the 120 URLs identified from our searches, 51 were duplicates, and 18 were excluded as they were either peer-reviewed articles, websites directing to video-sharing platforms or irrelevant to the research question. Fifty-one unique websites were included for analysis (Fig. 1). Full search results including reasons for exclusion and basic information about included websites can be found in supplementary file 3.

Fig. 1.

Flow chart.

For credibility, there was a 92.5 ​% agreement rate between independent reviewers before consensus. The date when the website was created and/or last updated was not mentioned in 45.1 ​% (23/51) of the websites. Disclosure of conflict of interest, sponsorship and funding source was not mentioned in 100 ​% (51/51) of the websites. Declaration of authorship was not mentioned in 29.4 ​% (15/51) of the websites. Presentation of a list of references was not done on 55 ​% (28/51) of the websites. Additionally, 19.6 ​% (10/51) of the websites were created to advertise products or services (commercial websites). Full details of the credibility findings for all websites can be found in supplementary file 4.

The mean Flesch reading ease score was 62 out of 100 (median, range ​= ​63, 42–88) and Flesch–Kincaid grade level was 8 (median, range ​= ​8, 3–12) across all websites. Full details of readability findings for each website can be found in supplementary file 5.

We found eight Australia-based websites, which had similar credibility and readability scores compared to the other 43 non-Australia-based websites (Table 1). Also, we found 10 websites advertising a product or service (commercial websites), which had similar credibility and readability compared to the other 41 websites not advertising a product or service (non-commercial websites) (Table 1).

Table 1.

Sensitivity analyses comparing Australia-based and non-Australia-based websites, and commercial and non-commercial websites.

	Australia-based (n ​= ​8)	Non-Australia-based (n ​= ​43)
Credibility, % (number of participants)
No currency of information	12.5 ​% (1/8)	28 ​% (12/43)
No disclosure of conflict of interest	100 ​% (8/8)	100 ​% (43/43)
No declaration of authorship	50 ​% (4/8)	25.5 ​% (11/43)
No presentation of a list of references	50 ​% (4/8)	56 ​% (24/43)
Readability, median (range)	63 (42–88)	62 (42–76)
	Commercial (n ​= ​10)	Non-commercial (n ​= ​41)
Credibility, % (number of participants)
No currency of information	20 ​% (2/10)	27 ​% (11/41)
No disclosure of conflict of interest	100 ​% (10/10)	100 ​% (41/41)
No declaration of authorship	60 ​% (6/10)	22 ​% (9/41)
No presentation of a list of references	70 ​% (7/10)	46 ​% (19/41)
Readability, median (range)	60 (44–70)	63 (42–88)

Six main thematic units emerged in the analysis of the URLs' texts: (i) What is knee crepitus? (ii) Is knee crepitus a problem? (iii) What is the cause of knee crepitus? (iv) How to manage knee crepitus? (v) What are the health consequences of having knee crepitus? (vi) Who should I seek if I have knee crepitus? The sub-themes raised from our reflective analysis are described in Table 2 as well as if they are appropriately supported or partially supported by evidence (accurate and partially accurate, respectively), if there is evidence opposing the sub-theme (inaccurate) or if there is no clear evidence (unable to determine) about the sub-theme. Fig. 2 shows the combination of the most frequent words used in the websites.

Table 2.

Qualitative synthesis and accuracy of the information.

What is knee crepitus?
Sub-theme	Quote	Accuracy	Explanation
Noise-related definition	“Crepitus describes a popping, clicking or crackling sound in a joint.” (URL05) “Another term commonly used for popping in the knee is “crepitus”, which essentially means a noisy joint, whether it be popping, clicking, cracking or snapping.” (URL14)		The definition of knee crepitus is not only based on noise but also on joint movement [11].
Movement-related definition	“You may hear occasional pops, snaps, and crackles when you bend or straighten your knees, or when you walk or go up or down stairs.” (URL01) “Crepitus is a symptom caused by vibrations produced by articular cartilage during the flexion-extension movement of the knee.” (URL16)		The definition of knee crepitus is based on noise and joint movement [11].
Is knee crepitus a problem?
Sub-theme	Quote	Accuracy	Explanation
Pain-free knee crepitus is not a problem	“If you have knee clicking but it doesn't cause you any pain, don't worry. It is usually entirely normal and nothing to worry about.” (URL14) “Crepitus is not a disease or condition — it's just something that happens in your body — and it isn't anything to worry about if you don't have other symptoms.” (URL74)		Knee crepitus is present in asymptomatic individuals, and it is not associated with pain in patients with patellofemoral pain [13,30].
Knee crepitus is a problem when combined with other symptoms	“When swelling, pain or both accompany knee popping it could indicate injury to the tissues — soft tissue or bone — surrounding the knee, referred to as pathological noise.” (URL52)		Knee crepitus is not associated with symptoms, but it is a diagnostic criterion for some knee conditions [8].
Knee crepitus is associated with symptoms severity	“If the crunching sound in your knee is accompanied by symptoms like pain and swelling, it could be signs of serious underlying issues.” (URL33)		Knee crepitus is not associated with symptoms in patients with patellofemoral pain [11].
What is the cause of knee crepitus?
Sub-theme	Quote	Accuracy	Explanation
Knee joint physiology	“It usually happens when air bubbles develop in body tissues, but it can also happen when the tendons or ligaments snap over the bony structures in the knee.” (URL16)		There is no evidence to support this information.
Intra-articular pathologies	“At other times, snapping sensations, cracking, and popping are symptoms of the conditions: ligament injuries, meniscus tear, osteoarthritis, and runner's knee.” (URL38) “The most common cause of frequent popping is osteoarthritis. However, other symptoms may point towards joint damage, bursitis, tenosynovitis, a bone fracture, gout, or arthritis.” (URL74)		There is no evidence to support this information.
Aging	“Don't ignore your noisy knees. They may be trying to tell you something, especially if you are over the age of 50.” (URL115) “This problem is not as age-related as it may seem” (URL117)		There is no evidence to support this information.
How to manage knee crepitus?
Sub-theme	Quote	Accuracy	Explanation
Exercises	“Exercise plays an important role in treating knee crepitus. Strengthening all the muscles around the knee is the single most important exercise for this condition.” (URL77)		There is no evidence to support this information.
Medication	“Many causes of joint popping and cracking improve with home remedies, like taking anti-inflammatory medications.” (URL05)		There is no evidence to support this information.
Passive treatments and rest	“Your doctor will talk with you about possible noninvasive treatments for crepitus, including bracing or splinting.” (URL05) “How do you treat knee crepitus – rest, ice, compression, and elevation are the first lines of treatment for this condition.” (URL100)		There is no evidence to support this information.
No treatment	“Most crepitus does not require treatment to fix, as there is nothing wrong with the knee in the first place.” (URL13)		There is no evidence to support this information.
What are the consequences of having knee crepitus?
Sub-theme	Quote	Accuracy	Explanation
Knee crepitus is a risk factor for knee osteoarthritis	“If crunching on your knee continues for long periods of time, it is possible for the cartilage in your knee to roughen. As a result, your risk of developing osteoarthritis can increase.” (URL33)		Knee crepitus has been reported to predict symptomatic knee osteoarthritis, but there is no evidence investigating the association between cartilage damage and knee crepitus [10].
Knee crepitus is not a risk factor for knee osteoarthritis	“There is no evidence to suggest that crepitus “wears the knee out” or causes arthritis.” (URL13)		Knee crepitus has been reported to predict symptomatic knee osteoarthritis, but there is no evidence investigating the association between cartilage damage and knee crepitus [10].
People with knee crepitus are likely to have knee pain	“This study suggests that if these people have noisy knees, they are at higher risk for developing pain within the next year compared to the people who do not have noisy knees.” (URL115)		Knee crepitus is present in asymptomatic individuals, and it is not associated with pain in patients with patellofemoral pain [13].
Knee crepitus is not related with knee pain	“Crepitus is not a predictor of pain and/or future knee function.” (URL13)		Knee crepitus is present in asymptomatic individuals, and it is not associated with pain in patients with patellofemoral pain [11].
Who should I seek if I have knee crepitus?
Sub-theme	Quote	Accuracy	Explanation
Medical doctor	“If you're worried about new sounds your knees are making, go see a doctor.” (URL108)		There is no evidence to support this information.
Physiotherapist	“You should consider seeking medical attention from a health professional, such as a physiotherapist.” (URL06)		There is no evidence to support this information.

- Accurate; - Partially accurate; - Inaccurate; - Unable to determine.

Fig. 2.

Combination of the most frequent words used in the websites.

4. Discussion

Our systematic appraisal of online information about knee crepitus indicates there was a large variation in the overall credibility and readability of content, with more than half of the websites lacking credible information. The main topics covered by websites were: (i) What is knee crepitus? (ii) Is knee crepitus a problem? (iii) What is the cause of knee crepitus? (iv) How to manage knee crepitus? (v) What are the health consequences of having knee crepitus? (vi) Who should I seek if I have knee crepitus? There were a few instances where the content provided was appropriately supported by evidence, but for most, we were unable to determine due to a lack of evidence to support the assumptions. Considering that people with clinical symptoms (e.g. crepitus) are increasingly seeking health information online [26], healthcare professionals should advise patients that online information may be inaccurate, and address any misinformed beliefs.

No website provided information about conflict of interest, and more than one-third did not disclose authorship details. These factors are considered important to the general population when determining the credibility of health information on the Internet [27]. Interestingly, patients may also judge the credibility of websites differently than the criteria used by our study. For example, a previous study reported that the perceived credibility of websites can be influenced by factors such as design, layout and inclusion of graphics [27]. This is concerning as well-designed websites can still present inaccurate information. As a guidance to readers seeking web-based sources, websites that referenced peer-reviewed scientific articles had more credible information. The readability of the texts was considered very easy and mostly intended for people with eight years of education. Only one in five of the websites exceeded the recommended grade level for the readability of health information, which is consistent with previous reports about knee osteoarthritis and low back pain [6,28].

Our qualitative synthesis indicated most websites focused on explaining what knee crepitus is and this was probably driven by a general public demand. Findings from a qualitative study exploring patient's beliefs about knee crepitus stated that health professionals cannot explain what crepitus is (e.g. “Nobody said anything about it even when they asked about the symptoms. I was just completely glossed over and ignored”) [9]. Overall, the websites presented accurate definitions of crepitus considering noise and movement-related definitions [11,29]. The causes and implications of knee crepitus were also highly discussed on the websites with a mix of accurate, inaccurate and unable-to-determine sub-themes raised from our analysis. Against the information reported by some websites, there is sound evidence supporting that crepitus is not associated with symptom severity in people with knee conditions [8,11,13,25]. Interestingly, a few websites reported that pain-free people should not be worried about crepitus as this is a normal sign, which is in accordance with reports that 99 ​% [30], 33.5 ​% [11], and 55 ​% [31] of asymptomatic people present knee crepitus.

Many websites suggested treatment options to reduce knee crepitus, including exercise, medication, passive treatments (e.g. brace, ice and splints) and rest. The literature around treatment options to reduce crepitus is scarce, apart from a few arthroscopy studies reporting the presence of crepitus pre- and post-surgery, other treatment options are non-existent. Clinical trials exploring interventions to reduce knee crepitus are warranted due to the absence of studies and how much this clinical sign matters to patients. However, websites should not be reporting any treatment recommendation to reduce knee crepitus until clinical trials are completed in this field [32]. Health professionals, such as medical doctors and physiotherapists are generally recommended to manage knee crepitus specially in commercial websites advertising products and services (e.g. appointments), these professionals currently have no source to find information about knee crepitus management.

The health consequences from having knee crepitus were also largely reported. We found conflicting reports among the websites, with instances where they stated knee crepitus is a risk factor for knee osteoarthritis and instances where it is not. In the literature, knee crepitus is considered one of the clinical findings that might be present to confirm a clinical diagnosis of osteoarthritis [18]. Two longitudinal studies exploring the Osteoarthritis Initiative database found that subjective knee crepitus can be used clinically to identify individuals at risk for symptomatic knee osteoarthritis [10], but it does not indicate they are more likely to undergo total knee replacement [25]. Obesity is a common risk factor for knee conditions and there was no mention about the relationship between obesity and knee crepitus [33,34]. Despite these previous reports, it is important to highlight that the long-term consequences of living with knee crepitus are unknown, particularly for pain-free and younger active populations.

We explored the internet using traditional search engines (Google and Bing), a large proportion of the population performs free searches in the browser linked to search engines and is directed to the websites we included. Due to feasibility, we could not identify all websites with information about crepitus, and we may have missed relevant websites that were not in the first 20 outputs of the search engines. However, people seeking health information are unlikely to view websites that do not appear on the first page of search results [6]. We did not assess the usability of the websites, it is possible that usability could influence how an individual interacts with the information [26]. Another limitation is that online information is constantly evolving; therefore, search results and findings from our study may change quickly. Also, searches will be influenced by factors such as geographical location, past searches/caches and whether the searcher is signed into personal accounts. Thus, people seeking information about knee crepitus may access and be exposed to different websites to those identified and analysed in our study.

There is a clear knowledge demand from the general public about knee crepitus. However, the literature available is still scarce, with important clinical questions yet to be clarified. Many topics identified in our qualitative analysis are not supported by research evidence (e.g. how to manage knee crepitus; what is the cause of knee crepitus?). Therefore, further studies exploring the role of knee crepitus are recommended, particularly with the inclusion of patients and health professionals when designing the research question.

5. Conclusion

Patients and clinicians should take caution when seeking web-based information about knee crepitus. Websites about knee crepitus lack credibility, present a large variation in readability outcomes, and there was no research evidence available to support most of their information.

Practical implications

• •Websites about knee crepitus present a large variation in readability and credibility outcomes.
• •More than half of the websites about knee crepitus do not present a list of peer-reviewed research articles to support their claims.
• •Many websites suggested treatment options to reduce knee crepitus, including exercise, medication, passive treatments (e.g. brace, ice and splints) and rest. However, none of them are supported by evidence.

Confirmation of ethical compliance

Ethical approval is not required for this study.

Funding information

This research did not receive any specific grant from funding agencies and the public, commercial, or not-for-profit sectors.

Declaration of competing interest

DOS is the communication editor of the Journal of Science and Medicine in Sport (JSAMS) Plus. FS and DAG are social media and digital content creators of JSAMS Plus.

Appendix A. Supplementary data

The following are the Supplementary data to this article.