Certified Athletic Trainers' Knowledge and Perceptions of Posttraumatic Osteoarthritis After Knee Injury

The Knowledge and Perceptions of Knee Osteoarthritis Survey (KPOAS) was administered online to 2000 ATs who were randomly sampled from all 10 National Athletic Trainers' Association (NATA) districts on April 1, 2014. We obtained e-mail addresses from the association, and the lead author (B.P.) sent potential respondents an invitation to complete the survey. The invitation included an embedded link to the survey, which was administered using a Qualtrics software platform (Research Suite; Qualtrics, LLC, Provo, UT). Reminder e-mails were sent 3, 5, and 7 days later to individuals who did not complete the survey. Following a 14-day period after the initial invitation to participate, data collection ended, and we sent nonrespondents no further correspondence. No incentives were provided to the survey participants. The Institutional Review Board at the University of North Carolina at Chapel Hill approved the online consent process and all KPOAS methods.

The KPOAS was created to evaluate the knowledge and perceptions of ATs pertaining to the risk of knee OA after ACL injury, ACL-R, meniscal injury, and meniscal surgery. Because no existing instrument was suitable, we adapted the KPOAS from a similar questionnaire that had been developed by 1 of the coauthors (K.L.B.) to evaluate the knowledge and perceptions of the risk of developing knee OA of patients undergoing ACL-R. The KPOAS comprises 2 sections regarding knowledge and perceptions: the risk of knee OA after knee injury or surgery and the therapeutic management of knee OA after knee injury or surgery. The KPOAS includes Likert-scale, visual-analog scale, yes/no, and multiple choice questions. Once the KPOAS was developed, 5 content experts reviewed it for content validity, and then a convenience sample of 15 ATs completed a pilot version. We used the pilot survey feedback to edit the instrument for face validity and readability. Because the KPOAS contents were designed to be examined at the item level rather than as a higher-order knowledge variable, no specific factor analytic procedures were conducted. We provide the KPOAS in the Appendix.

The first question in this section asks participants to choose the option that best defines OA and offers 7 options: (1) a condition in which the bones become fragile and brittle, causing them to break more easily, (2) a name given to a group of symptoms marked by generalized pain and muscle stiffness, (3) a condition that affects a number of structures around the joint and is characterized by the breakdown of the joint's cartilage, (4) an autoimmune disease that causes pain and swelling of the joints, (5) all of the above, (6) none of the above, and (7) do not know. Then respondents are asked where and how they learned about OA (ie, personal experience, other clinicians, formal coursework, and continuing education). The next portion consists of 16 Likert-scale items (strongly disagree, disagree, do not know, agree, strongly agree) designed to determine the knowledge of risk factors associated with knee OA and the perception of OA as a major health care concern. Three of the 16 items are incorrect and would be correctly identified as strongly disagree but were included in the KPOAS as distractor options (ie, cold weather, high alcohol intake, being male). Four visual-analog scale questions asked the percentage of patients who were ACL deficient (ACL-D) or had undergone ACL-R, partial meniscectomy, or full meniscectomy who would develop knee OA in the first, second, third, and fourth decades after injury; respondents choose a number between 1% and 100% indicating the percentage of patients who would develop knee OA at each time point. By selecting yes or no, participants state if they are aware of a condition termed PTOA. This question was placed last in an effort to avoid biasing respondents regarding the purpose of the survey.

This section of the KPOAS consists of dichotomous yes or no questions to evaluate if participating ATs (1) treat patients with ACL or meniscal injury, (2) explain the risk of developing knee OA after acute knee injury, and (3) provide strategies for preventing knee OA to patients after acute knee injury or surgery. The first of 7 Likert-scale items asks, “How important is it to you that your patients undertake strategies to prevent osteoarthritis from developing in their knee following ACL or meniscus injury?” Answer options are not at all important, slightly important, moderately important, very important, and extremely important. The next 6 questions were designed to determine how likely (very unlikely, unlikely, unsure, likely, or very likely) ATs are to prescribe specific interventions (strengthening, maintaining an appropriate body weight, supportive knee braces, shoe orthotics, or wearing specially designed shoes) that may reduce the risk of knee OA after an ACL or meniscal injury. We decided to collect data regarding these specific interventions based upon recommendations from previous researchers.¹⁶  Participant demographic information was collected after all questions pertaining to OA risk and the therapeutic management of OA were answered.

Data from all participants who submitted the survey were electronically converted from the Qualtrics platform to SPSS (version 17.0; IBM Corporation, Armonk, NY) for data analysis. We evaluated the ATs' knowledge and perceptions of knee OA as well as the potential use of treatment strategies by calculating and reporting the means, standard deviations, frequencies of responses to the survey questions in each section, and percentages for descriptive participants' demographic information. Respondents were allowed to skip any question they did not feel comfortable answering. Therefore, the number of ATs who responded to each question differed, and the sample size is reported for each question. We calculated the percentage of participants who believed that the incidence of knee OA would be higher in patients who underwent an ACL-R or meniscal repair compared with those who remained ACL-D or underwent a meniscectomy, respectively. Individual dependent-samples t tests were used to determine if the ATs' estimated percentages of patients who would develop OA after ACL-R versus ACL-D differed in each decade postinjury. Similar dependent-samples t tests were conducted to determine if the ATs' estimated percentages of patients who would develop OA after a meniscal repair versus meniscectomy differed in each decade postsurgery. Given the exploratory nature of the study, we set the α levels a priori at .05 for the dependent-samples t tests.

In addition to reporting frequencies, percentages, and descriptive statistics, we performed χ² analyses to understand the interaction between the knowledge of PTOA and the use of therapeutic strategies to mitigate the risk of knee OA development in patients with ACL or meniscal injury or surgery. We used separate Pearson χ² models to examine if the observed responses differed from the expected responses for 4 analyses: (1) Were ATs who reported specific overall treatment practices more likely to be aware of the condition of PTOA? (2) Did the relative importance placed on undertaking strategies to prevent knee OA after ACL injury and reconstruction or meniscal injury or surgery differ between respondents who were and were not aware of the condition of PTOA? (3) Were responses regarding the likelihood of using various treatments to mitigate the risk of developing knee OA different between respondents who were and were not aware of the condition of PTOA? (4) Did the percentage of respondents who were aware of the condition of PTOA differ by years of experience as an AT (1–5, 6–10, 11–15, 16–20, >20 years)? We used a Bonferroni correction for the χ² analyses evaluating the importance placed on undertaking strategies, likelihood of using different treatments, and reported years of experience as an AT based upon the number of potential responses (0.05/5) and set P at ≤.01. If a cell contained fewer than 5 samples, we used an α level associated with a 2-sided Fisher exact test to determine differences between cells.

The surveywide response rate was 21.9% (437/2000). The sample responding to all questions was 347 ATs (17.4%). Full demographic information was provided by 369 ATs (40.5% male and 44.6% female; Table 1). Fifty percent reported having completed a master's degree. Certification as a Strength and Conditioning Specialist was the most frequently reported credential (11%) after certified AT. Only 11.2% of the sample reported an athletic training clinic as the primary place of employment; physical therapy clinics (40.7%) and physician's offices (31.1%) were the most commonly reported places of employment (Table 1). Participants most often reported having more than 20 years of experience as an AT (20.6%). Of those sampled, 14%, 29.9%, and 22.8%, respectively, had personally sustained an ACL injury or a meniscal tear or had been diagnosed with OA. Of the 84 respondents who reported being diagnosed with OA, 67 (77%; 15.3% of the total sample) specifically reported it affected the knee joint.

The majority of the sample (84.7%) correctly identified OA as a condition that affects a number of structures around the joint and is characterized by the breakdown of the joint's cartilage. Most reported that they learned about OA from personal experience (n = 241, 55.1%), other clinicians (n = 235, 53.8%), formal coursework (n = 277, 51.9%), and continuing education (n = 220, 50.3%). A combined 61.2% (n = 267) of the sample either strongly agreed or agreed that a patient having knee OA would be a major health concern, which was similar to the percentage of ATs who did have knowledge of the condition of PTOA (n = 225, 60.3%; Table 2). The majority of respondents either strongly agreed or agreed that certain factors would increase the risk of developing OA, including (1) full meniscectomy (n = 431, 98.9%), (2) being overweight (n = 420, 96.2%), (3) meniscal tear (n = 415, 95.4%), (4) altered biomechanics after an acute injury (n = 394, 90.3%), (5) an ACL tear (n = 393, 90.2%), (6) partial meniscectomy (n = 392, 90.1%), (7) persistent strength deficits after an acute injury (n = 368, 84.6%), and (8) smoking (n = 274, 62.7%; Table 3). There was no majority consensus (agree, disagree, or do not know) as to whether the following factors increased the risk of OA development: meniscal repair, male sex, or having a relative with knee OA (Table 2). A total of 51.3% (n = 224) of the respondents either strongly agreed or agreed that having strong quadriceps reduces the risk of developing knee OA, whereas 51.4% (n = 194) of respondents strongly disagreed or disagreed that ACL-R reduces the risk of developing knee OA (Table 2).

A greater number of participants hypothesized that 20-year-old ACL-injured patients would have a higher prevalence of OA at ages 30, 40, 50, and 60 years if they remained ACL-D versus undergoing ACL-R (P = <.001; Table 3). At all time points, the majority of ATs (30 years = 53%, 40 years = 56%, 50 years = 56%, and 60 years = 56%) believed that a higher percentage of ACL-D patients would develop knee OA compared with ACL-R patients. Similarly, more respondents hypothesized that 20-year-old patients with meniscal injury would have a higher prevalence of OA at ages 30, 40, 50, and 60 years if they underwent meniscectomy compared with meniscal repair (P = <.001; Table 3). At all time points, the majority of participants (30 years = 66%, 40 years = 66%, 50 years = 66%, and 60 years = 67%) believed that a higher percentage of meniscectomy patients would develop knee OA than patients who underwent a meniscal repair.

Most respondents reported treating patients with ACL (n = 400, 92.8%) and meniscal injuries (n = 386, 88.3%). The majority rated the importance of undertaking strategies to prevent OA from developing after ACL injury or ACL-R (n = 296, 73.8%) and meniscal injury or surgery (n = 317, 74.7%) as extremely important or somewhat important (Table 4). Most reported it was very likely or likely that they would recommend the following interventions to patients after such an injury or surgery: (1) ongoing regular lower limb muscle strengthening/retraining (n = 396, 98.4%), (2) maintaining an appropriate body weight (n = 365, 97.1%), (3) wearing a supportive knee brace during physical activities and sports (n = 285, 76%), and (4) using shoe orthotics on a daily basis (n = 286, 76.5%; Table 4). There was no treatment for which the majority of ATs reported the usage as being very unlikely or unlikely. Although most believed that it was appropriate to explain the risk of OA to patients after ACL or meniscal injury (n = 419, 96.8%), a lower percentage actually reported explaining the risk of OA to patients after an ACL (n = 277, 70.8%) or meniscal (n = 308, 80.6%) injury or surgery. A similar percentage reported providing information to patients about strategies to lower their risk of developing knee OA after an ACL (n = 277, 70.8%) or meniscal (n = 294, 78.4%) injury or surgery.

We found that 60.5% of ATs (n = 210) who treated ACL-injured patients and the same percentage who treated meniscus-injured patients did have knowledge of PTOA. Participants who reported having knowledge of PTOA were more likely to explain the risk of developing OA after ACL injury (P < .001) and meniscal injury (P = .001) and to provide patients with strategies to lower the risk of OA after ACL (P < .001) and meniscal injury (P < .001; Table 5) compared with those who did not have such knowledge. No differences were evident in the likelihood of ATs choosing different treatments based on their knowledge of PTOA (Table 6). As years of credentialed experience increased, progressively higher percentages of ATs had knowledge of PTOA (P = .009). The ATs with more than 20 years of clinical experience (74.4%) had the highest percentage of respondents with knowledge of PTOA, whereas those with 1–5 or 11–15 years of clinical experience had the lowest percentage of respondents with such knowledge (52.9%). Of the respondents with 6–10 and 16–20 years of experience, 54.4% and 68.8%, respectively, reported having knowledge of PTOA.

We conducted post hoc Pearson χ² analyses to investigate the results of the planned primary analyses. A previous history of an ACL injury (χ² = 0.12, P = .73) or a meniscal injury (χ² = 1.51, P = .73) did not increase the percentage of people who were aware of PTOA. A greater percentage of respondents reported being diagnosed with arthritis and aware of the condition of PTOA (70.8%) compared with those who had never been diagnosed with arthritis and were aware of the condition of PTOA (56.7%, χ² = 5.97, P = .02). Additionally, the ability to correctly define OA was not affected by years of experience as an AT (χ² = 25.25, P = .19).

We surveyed a large number of ATs about their knowledge and perceptions of OA, as well as their educational strategies and treatment recommendations for patients with PTOA. We found that ATs were knowledgeable about the general condition of OA. Although most knew that ACL and meniscal injury or related surgery increased the risk of developing knee OA, more than one-third (approximately 40%) were not familiar with the condition labeled PTOA. The unfamiliarity with the term PTOA may be evidence that a large proportion of ATs are not knowledgeable about the specific factors that differ between PTOA and other OA phenotypes that develop via idiopathic origins. Data from the current study suggest that ATs may not be aware of the high percentage of patients who develop PTOA after traumatic injury and the rapid onset of metabolic and structural osteoarthritic changes after an acute injury. Knowledge of risk factors for knee OA was generally good, but the ATs seemed to overestimate the prophylactic benefit of ACL-R in decreasing the risk of PTOA; current evidence⁸  indicates that ACL-R does not decrease the risk of OA. Athletic trainers believed that undertaking strategies to prevent the onset of knee OA in patients with knee injuries was important, and they were likely to recommend a number of interventions. Differences in the likelihood of explaining the risk of developing knee OA and providing treatment recommendations were noted when we compared ATs with and without knowledge of PTOA. The ability to correctly identify the definition of OA was not affected by years of experience as an AT, yet knowledge of PTOA seemed to be related to more years of clinical experience. A history of a personal ACL or meniscal injury did not affect the percentage of respondents who were aware of the condition of PTOA; however, a greater percentage of those who had been diagnosed with arthritis were aware of PTOA than those who had not been diagnosed with arthritis. Athletic trainers with a diagnosis of arthritis may be more aware of PTOA, but respondents who have had a traumatic knee (ACL or meniscal) injury and may be at higher risk for developing PTOA themselves were not more aware of PTOA than those who had not sustained such an injury.

One of the 5 most debilitating conditions in the United States,¹  OA is also the fourth most common cause of hospitalization.¹⁷  Knee OA may increase chronic pain and decrease physical activity, leading to inactivity-related conditions and comorbidities that increase mortality rates.¹⁸  Knee OA has been linked to decreased physical activity^19,20  and higher risks of cardiovascular disease, obesity, diabetes,¹⁸  and depression.²¹  Although PTOA patients account for only 12% of the total population with OA,²²  because they develop the condition earlier in life, they may live more years with disability than OA patients who develop the disease later in life. Additionally, people with PTOA have higher rates of knee arthroplasty failure and revision.^23,24  More than 90% of the ATs sampled agreed that ACL or meniscal injury or surgery increases the risk of developing knee OA, but only about 60% agreed that knee OA is a serious health concern (Table 2). There may be a disconnection between the knowledge of the risk of PTOA and the perception of the seriousness of developing OA. Educating ATs about the possibility that their patients with ACL or meniscal injury will develop OA may help advance treatment strategies and maximize long-term joint health.

Athletic trainers who responded to the survey provided appropriate answers to questions about their specific knowledge of OA, suggesting a general working knowledge of the disease. The literature supports the answers provided by the majority of respondents who strongly agreed or agreed that full meniscectomy (98.9%), meniscal tear (95.4%), ACL injury (90.2%), or partial meniscectomy (90.1%) increases the risk of knee OA (Table 2).^8,25–27  Athletic trainers estimated the risk of OA as higher at all hypothetical time points (decades 1, 2, 3, and 4 after ACL injury) for patients who remained ACL-D compared with patients who underwent ACL-R (Table 5, P < .001). However, current research⁸  does not support the notion that ACL-R decreases the risk of knee OA compared with patients who remain ACL-D. Investigators⁸  have observed a 6% absolute increase in the risk of developing OA after ACL-R compared with patients who remained ACL-D. We can speculate that ACL-R patients may engage in more physical activity or activities that require more load on the knee than ACL-D patients and, therefore, develop PTOA at a similar rate to ACL-D patients who may have less physically active lifestyles. Although the authors of a previous systematic review⁸  reported similarly decreased physical activity levels in both ACL-R and ACL-D patients postinjury, more research is necessary to determine the mechanisms that lead to a similar prevalence of PTOA among ACL-R and ACL-D patients. Similarly, ATs estimated the risk of OA as higher after meniscectomy than after meniscal repair (Table 3). The comparative risk of OA after meniscectomy versus meniscal repair is not yet known, but the amount of meniscus removed may be related to the progression of early joint degeneration.²⁸  Athletic trainers may underestimate the overall risk and rapid onset of OA after knee injury. Conversely, they potentially overestimate the prophylactic capability of ACL-R to decrease the risk of OA. This could result in their giving patients inaccurate information and lead patients considering an ACL-R to discount OA as a future concern.

The majority of ATs indicated that it was extremely or somewhat important to undertake strategies to prevent OA after ACL injury, ACL-R, meniscal injury, and meniscal surgery (Table 4). Athletic trainers who had knowledge of PTOA were more likely to indicate that undertaking strategies to prevent OA was extremely important for patients than those who did not have knowledge of PTOA (Table 6). Although 96.8% of ATs indicated that it was appropriate to explain the risk of OA to patients after an ACL injury or meniscal injury, only 70.6% and 80.6%, respectively, provided such information to patients. A higher percentage of ATs who had knowledge of PTOA explained the risk of OA after ACL or meniscal injury compared with those who did not have such knowledge. At this point, it is unclear why ATs who believe that it is appropriate to explain the risk of developing OA after ACL or meniscal injury fail to do so. Perhaps some ATs are uncomfortable providing information about the OA risk to their patients, do not feel equipped with effective strategies to explain the risk, or rely on other health care professionals to deliver that information. It is important to identify the barriers that prevent ATs from delivering information about the PTOA risk to patients so that strategies can be developed to overcome these barriers and more effectively engage patients. We are careful not to conclude that educating ATs will change clinical practice regarding interventions to limit PTOA, but improving their knowledge about PTOA may increase the likelihood that they will provide information regarding the risk of OA to patients. For the most part, knowledge of PTOA did not affect the likelihood of using the specific types of interventions that we provided as options, indicating that, regardless of their knowledge of the disease, clinicians use muscle strengthening, knee braces, orthotics, specialized footwear, and maintenance of a proper body weight to the same extent to prevent OA (Table 6).

Improving knowledge of PTOA among ATs may be a critical step in continuing to mitigate the risk of developing OA after acute injury. It is important to train clinicians to be attentive and responsive to the rapid advances in knowledge about the pathogenesis and the best management of PTOA that are generated by new research efforts. Educating ATs will likely lead to patients, caregivers (eg, parents), spouses, and coaches who are better informed about the future risk of developing a chronic disease and methods to maintain physical function over a lifespan. Athletic trainers most commonly reported learning about OA from personal experience (55.1%) or other clinicians (53.8%), yet only 22% reported actually being diagnosed with OA (Table 1). Consequently, a higher percentage of ATs who were knowledgeable about PTOA were in the cohort of clinicians who reported more than 20 years of clinical experience (74.4%) than in cohorts with less clinical experience. Athletic trainers with more years of practice may have experienced previous patients' reporting early-onset OA after a knee injury and be more knowledgeable about the negative long-term consequences of traumatic joint injury. Slightly more than 50% of respondents reported learning about OA from formal coursework (51.9%) or through continuing education (50.3%). Only 52.9% who had 1 to 5 years of clinical experience reported knowledge of PTOA, suggesting that efforts to integrate data about PTOA into undergraduate and graduate athletic training education coursework may improve the ability of novice ATs to treat patients at risk for PTOA. Our post hoc analyses did not demonstrate that the number of years of experience as a certified AT influenced the respondents' ability to correctly identify the definition of OA. Therefore, the deficiency may be in awareness about the pathogenesis and risk of the specific PTOA phenotype rather than knowledge of OA in general among less experienced ATs. Additionally, the largest cohort of ATs is between 22 and 37 years of age²⁹ ; our sample may have been slightly older, yet the age of our respondents was similar to the age of ATs who responded to online surveys in recently published research.^30,31  Although it is important to provide ATs with a better understanding of the risk and the factors related to the development of PTOA, it is also important to develop multimodal mechanisms that provide ATs with the most current information and innovative methods for decreasing the risk of PTOA in patients. Education about the long-term risks to joint health may need to be integrated into accredited athletic training education programs and the most current information disseminated to practicing ATs. Special interest groups within the NATA who are concerned about joint injury and disease should develop programs and online materials to inform ATs about PTOA. Furthermore, other organizations concerned with mitigating the negative effects of OA in the United States may educate ATs about PTOA as a way to improve early intervention and decrease the risk of OA after joint injury.

Despite the important implications of our findings, we acknowledge study limitations, which might inform future research and clinical efforts in this area. We sent our survey to a relatively large sample size (n = 2000), but only a small percentage responded (21.9%). Our sample contained more women than men, which is similar to membership trends reported by the NATA. However, compared with the NATA membership, a smaller percentage of respondents indicated that they were nonwhite or a minority race (12.4% versus 2.1%, respectively; Table 1).³²  Approximately 40% of the ATs stated that they worked in a physical therapy clinic and 31% worked in a physician's office, percentages that are much higher those reported by the NATA (18.4% and 4.2%, respectively).³²  It is possible that the ATs who responded to the study were interested in knee injury and more knowledgeable about it than those who did not reply, thereby leading to response bias. An additional study limitation involves the item-level interpretation of the KPOAS, which restricts our ability to examine a higher-order PTOA knowledge construct and its associated psychometric properties. Such efforts, though beyond the scope of the current study, would enhance our understanding of the construct validity of the KPOAS.

It is also possible that, due to the nature of the questions, respondents presumed a link between knee injury and OA, which may have affected how they answered the questions as they continued through the questionnaire. In case participants ended the survey prematurely, we elected to place the demographic questions at the end of the survey, which allowed us to collect the data pertaining to the knowledge of OA risk and treatment first. Unfortunately, we did not gain demographic information from approximately 3.5% of the ATs who volunteered. Despite these limitations, our study provides a solid baseline with which to better understand and influence ATs' knowledge and best practices regarding knee injury and OA.

Although ATs were knowledgeable about the general condition of OA and most knew that ACL and meniscal injury increase the risk of developing knee OA, we found that more than one-third of the sample (approximately 40%) were not familiar with the condition when labeled as PTOA. Additionally, ATs seemed to overestimate the ability of ACL-R and current rehabilitative methods to decrease the risk of OA. We did note that ATs who were knowledgeable about PTOA were more likely to explain the risk of OA and provide patients with strategies that may help prevent OA after traumatic knee injury. Improving the knowledge about the risk of PTOA among ATs may be an initial step in improving treatment strategies to mitigate the risk of developing PTOA after acute knee injury.