Patient-Reported Outcome Measures in Sports Medicine: A Concise Resource for Clinicians and Researchers

To provide ATs with a concise guide to PROMs, we reviewed the instruments reported by ATs who used PROMs in routine practice. In a survey study by Lam et al,¹⁵  ATs who used PROMs on a routine basis were asked to identify the PROMs they used for patient care and research purposes. Based on the responses of 370 ATs who routinely used PROMs in care, 78 unique PROMs were endorsed and identified. We included PROMs in this report if at least 10% of the ATs endorsed their use in the study by Lam et al.¹⁵ 

Using the list of PROMs, we conducted a 2-phase literature search. First, we searched the literature with a focus on instrument development and establishment of the psychometric properties of each PROM. We completed 4 searches for each PROM using its name and the following key words: development, validity, reliability, responsiveness. For example, we performed these searches for the FAAM: (1) Foot and Ankle Ability Measure AND development, (2) Foot and Ankle Ability Measure AND validity, (3) Foot and Ankle Ability Measure AND reliability, and (4) Foot and Ankle Ability Measure AND responsiveness. For the second phase, we searched the literature for the use of the PROM specifically among athletes. For this phase, we used the PROM name in combination with 1 of 2 key words (athlet*, sport) in separate searches: for instance, (1) Foot and Ankle Ability Measure” AND athlet* and (2) Foot and Ankle Ability Measure AND sport. This process was repeated for each PROM.

We extracted data from the available literature to summarize the instrument essentials (ie, instrument development, reliability, validity, responsiveness and interpretability, and precision) and clinical utility (ie, acceptability [number of items, time to complete, readability, comfort-level concerns], feasibility [ease of use, role of clinician, time to score, costs associated with use], and appropriateness [intended patient populations, demonstrated use for other patient populations, global purpose of use]) of each PROM included in this report.

For instrument acceptability, we also assessed the readability of each PROM. Readability is important for all patients but particularly for patients who are minors, such as secondary school and youth athletes, or nonnative English speakers.^27,28  For this review, readability was represented by the Flesch-Kincaid reading grade level. To calculate the Flesch-Kincaid reading grade level, we used Word for Mac (version 16.15; Microsoft Corp, Redmond, WA). In short, each PROM was imported into Microsoft Word and its unformatted text was analyzed with the embedded formula to provide a reading grade level for the measure.

For instrument appropriateness, we also summarized the ICF health domains and HRQOL dimensions captured by each PROM using a consensus process described in a previous study.²⁹  In brief, the consensus process required each research team member (n = 3, all of whom had expertise in clinical outcomes assessment [eg, teaching, presentation, and research experience in clinical outcomes assessment]) to review the included PROMs independently and classify each PROM question within 1 ICF health domain²⁰  and 1 HRQOL dimension.^21,22  After performing independent reviews of all PROMs and initial classification of questions according to ICF health domains and HRQOL dimensions, the raters met as a group to compare their classifications. Discrepancies in classifications were discussed, and a final classification was determined by group consensus.

Descriptions of the specific ICF health domains and HRQOL dimensions used for this study were detailed in a previous investigation.²⁹  In brief, for the ICF health domains, raters classified each question in one of the following domains: health condition, body structure and function, activity, participation, environmental factors, or personal factors.^20,29  When necessary, the raters were able to consult the ICF Web site (apps.who.int/classifications/icfbrowser/) during the review process for guidance in categorizing ICF health domains. For the HRQOL dimensions, each item was classified in one of the following areas: physiological (ie, impairments such as pain and swelling), physical (ie, ability to perform activities and attributes such as mobility and performance), psychological (ie, emotional well-being, including happiness and sadness), spiritual (ie, value of religious beliefs and practices), social (ie, interactions with family and friends), or economic (ie, financial status and burden) functioning.^21,22,29 

Based on the findings of Lam et al,¹⁵  a total of 17 region-specific, 6 generic, and 3 single-item PROMs were endorsed by at least 10% of the ATs who used PROMs and thus were reviewed for this report. For region-specific PROMs, 11 lower extremity–specific (3 foot and ankle, 3 knee, 3 hip, 2 back) and 10 upper extremity–specific (3 shoulder-elbow, 3 wrist-hand, 1 neck, 3 head) instruments were studied (Table 1). Four PROMs were identified for use in multiple body regions: the Lower Extremity Functional Scale (knee and hip), Disabilities of the Arm, Shoulder and Hand (DASH; wrist-hand and shoulder-elbow), Quick-DASH (wrist-hand and shoulder-elbow), and Upper Extremity Functional Scale (wrist-hand and shoulder-elbow).

As a result, a total of 26 unique PROMs (10 lower extremity region specific, 7 upper extremity region specific, 6 generic, and 3 single item) were evaluated in this review. Consistent with Lam et al,¹⁵  we classified the PSFS as a single-item measure because it is neither a specific nor a generic measure. Table 1 provides a general summary of the instrument essentials and clinical utility of each included PROM for quick reference. More detailed summaries of the instrument essentials, including specific measurement property values, of lower extremity–specific, upper extremity–specific, and generic and single-item measures can be found in Tables 2, 3, and 4, respectively. Summaries of considerations for clinical utility can be found in Tables 5, 6, and 7.

Of the 10 lower and 7 upper extremity region-specific PROMs, all (100.0%, 17 of 17) were associated with the appropriate instrument essentials, with a reported systematic development process and evidence of reliability and validity (Tables 2 and 3). In addition, responsiveness values were reported for almost all region-specific PROMs (88.2%, 15 of 17), with the exception of the American Academy of Orthopedic Surgeons Foot and Ankle Questionnaire and Abbreviated Profiles of Mood States. The precision of the PROMs varied within and among instruments, with response scales including some combination of binary, modified visual analog scale, 3- to 7-point adjectival, and 5-point Likert-scale responses.

In terms of clinical utility, the region-specific PROMs also generally demonstrated appropriate acceptability (Tables 5 and 6). Patient completion time was estimated as less than 10 minutes for almost all of the PROMs (15 of 17, 88.2%), with the expectation that many could be completed in 5 minutes or less (11 of 17, 64.7%). Readability of the measures ranged from fourth to 10th grade (Table 5) and third to sixth grade (Table 6) for the lower extremity and upper extremity PROMs, respectively. Most PROMs (13 of 17, 76.6%) had an estimated reading level of seventh grade or below. The region-specific PROMs also demonstrated good feasibility, with none requiring (1) special training to understand the administration process, (2) a clinician to complete the questions, or (3) clinician supervision of the patient during completion. Although 3 instruments (17.6%) required a user agreement, only 1 instrument, the Shortened Headache Impact Test, required paid access for use. In addition, the clinician burden was relatively low, with the time to score each measure estimated at ≤5 minutes. In terms of appropriateness, most appeared relevant to the types of conditions or areas of health effect that are important to athletes. Further, although the majority of the region-specific PROMs appeared to address items of importance to athletes, most were not developed specifically for high-functioning athletic populations (94.1%, 16 of 17). From an ICF health domain perspective, the region-specific PROMs generally captured the body structure and function (39.1%, 163 of 417 items) and activity (45.1%, 188 of 417 items) domains. Very few of the items on the region-specific instruments were related to the participation (13.2%, 55 of 417 items) or environmental factors (2.6%, 11 of 417 items) domain, and none included questions related to the health condition domain. From an HRQOL dimension perspective, the region-specific PROM instruments included questions that predominantly evaluated the physical (54.7%, 228 of 417 items) and physiological (23.7%, 99 of 417 items) dimensions. The psychological (12.7%, 53 of 417) and social (8.9%, 37 of 417 items) dimensions were captured less frequently, and none of the region-specific measures addressed the spiritual or economic dimension.

We reviewed 6 generic PROMs. All were developed using a systematic process, had evidence of reliability and validity, and had established responsiveness values in some populations (Table 4). Similar to the region-specific PROMs, the precision of the generic PROMs varied within and among instruments, with questions requiring binary, 3- to 11-point adjectival, and 5- to 15-point Likert responses.

In general, most generic PROMs (4 of 6, 66.7%) demonstrated good acceptability and feasibility, including appropriate patient completion time (less than 5 minutes), no comfort-level concerns, and limited clinician burden associated with the Disablement in the Physically Active (DPA) scale, Pediatric Quality of Life, Medical Outcomes Study 36-Item Short Form Health Survey, and Short Form 12 (Table 7). Of note, the Musculoskeletal Function Assessment (MFA) and Short MFA both consist of more items (110 and 46, respectively) and, thus, require more time to complete (15 and 5–10 minutes, respectively) relative to the other generic PROMs. In addition, the MFA and Short MFA also include items with potential comfort-level items (ie, Has your sexual life changed? Do you enjoy sex less? How much difficulty are you having with sexual activity?) for patients. The readability of the included generic PROMs ranged from second to 10th grade, with 77.8% (7 of 9) estimated at sixth grade or below (Table 7). From an ICF health domain perspective, the generic PROMs generally captured the body structure and function (35.0%, 85 of 243 items), activity (35.0%, 85 of 243 items), and participation (30.0%, 73 of 243 items) domains. None of the items on the generic instruments were related to the health condition or environmental factors domain. From an HRQOL dimension perspective, the generic instruments included questions that predominantly evaluated the physical (40.7%, 99 of 243 items), physiological (25.1%, 61 of 243 items), and social (23.9%, 58 of 243 items) dimensions of health. The psychological (10.3%, 25 of 243) dimension was captured less frequently, and none of the region-specific measures addressed the spiritual or economic dimension.

The Numeric Pain Rating Scale (NPRS), Global Rating of Change (GROC), and PSFS have established reliability, validity, and responsiveness in some populations; however, only the GROC involved a systematic development process (Table 4). The NPRS and PSFS use 10- and 5-point adjectival scales, respectively, whereas the GROC was developed as a 15-point Likert-type scale.

The single-item measures appear to have good acceptability and feasibility with short times for patient completion (<3 minutes) and low clinician burden (<1 minute; Table 7). The NPRS captures the body structure and function domain and the physiological HRQOL dimension, and the PSFS assesses the activity domain and physical HRQOL dimension; however, the ICF health domain and HRQOL dimension captured by the GROC vary, as they depend on how patients perceive their condition or injury and the subsequent frame of reference when reflecting on and answering the question.

To our knowledge, this is the first report to critically review and summarize the instrument essentials and clinical utility of generic, specific, and single-item PROMs that are used in athletic health care. In addition, we provided a summary of the ICF health domains and HRQOL dimensions that questions within each PROM addressed to offer insight into their use when delivering patient-centered care. Overall, we aimed to provide a helpful, concise resource for ATs to consult when selecting and implementing PROMs.

In general, the PROMs studied in this commentary demonstrated appropriate instrument essentials, with almost all having a systematic development process and acceptable psychometric properties including reliability, validity, and responsiveness. However, it is important to note that only a few of the instruments were specifically designed to evaluate aspects of disablement and health among highly functional patients, such as athletes. For example, of the instruments reviewed, only the DPA^77,78  and the Abbreviated Profile of Mood States⁷⁴  were developed with athletes as the intended population. Further, much of the research to date related to the instrument measurement properties of generic, specific, and single-item PROMs has been conducted in populations other than a highly functional patient population such as athletes. This finding is a concern when considering the validity of the instruments for use in athletic health care.

Other PROMs were designed for the athletic population, such as the Kerlan-Jobe Orthopaedic Clinic overhead athlete score,¹⁴²  the Functional Arm Scale for Throwers,^143,144  the Athlete Fear Avoidance Questionnaire,¹⁴⁵  and the Swimmer's Functional Pain Scale.¹⁴⁶  Yet previous research¹⁵  indicated that fewer than 10% of ATs routinely used these instruments. Thus, they were not included in this report. However, even though many of the included PROMs were developed for more general populations, these patients often presented with injuries similar to those sustained by athletes.^35,53  Ideally, measurement properties such as reliability, validity, and responsiveness should be established for the intended population.¹⁹  Because evidence^147–149  suggested that the HRQOL of highly functional patients is different than that of the general population, future work is needed to establish the validity, reliability, and responsiveness of the PROMs most commonly used in athletes.

The PROMs included in this report appear appropriate, acceptable, and feasible for use in athletic health care. Considering readability specifically, the general guidance was that the calculated reading grade level be 2 reading levels below a patient's actual grade level.^27,28  For example, a patient in the ninth grade should be administered a PROM with a reading level of seventh grade or lower. Of the reviewed PROMs, the vast majority (20 of 26, 76.2%) had an estimated reading level of seventh grade or lower, suggesting that they would likely be appropriate for adult and adolescent patients. However, it is important to note that a patient's grade level may not necessarily align with his or her actual reading level (eg, students of English as a second language); clinicians should take this into account when selecting a PROM.

When we assessed the ICF health domains and HRQOL dimensions of health captured by the reviewed PROMs, it was not surprising that many of the instruments emphasized specific aspects of health. Most instruments include questions that evaluate the ICF health domains of body structure and function and the HRQOL dimensions of physiological and physical functioning. For example, the Lower Extremity Functional Scale and the FAAM are region-specific PROMs that focus solely on functional ability. Using PROMs that evaluate physical function in athletic health care is appropriate because highly functional patients often focus on maintaining or regaining high levels of physical function to perform activities in daily life and sports. For example, a common goal of athletes is to restore function to compete in their sports and fulfill their role as an athlete. Instruments that evaluate function allow ATs to better direct rehabilitation to meet these performance and role goals (ie, participation domain of the ICF). Even though regaining function is a common goal of athletes, other ICF health domains and HRQOL dimensions may warrant attention.¹⁹ 

Information related to body structures and functions, such as range of motion and strength, is helpful for clinicians to obtain a more complete understanding of the status of tissue healing, which may support treatment decisions to promote continued recovery.²⁰  An equally important area of health to evaluate is participation. However, the participation domain was not a frequent component of the PROMs included in this review. Participation reflects the areas of health that many patients care most about because it relates to the ability to complete necessary or desired life roles, such as athlete, friend, student, parent, or employee.²⁰  When selecting PROMs, ATs should consider whether the patient case warrants evaluation of the participation domain, particularly because athletes often have a strong identity grounded in being an athlete. The effect of identity loss due to injury and removal from sport may be an important focus when managing and coordinating care for a patient. In general, generic instruments include more questions that capture participation than specific instruments because they are designed to assess health on a more global level. However, some of the region-specific measures, such as the DASH,⁵³  the Dizziness Handicap Inventory,⁷⁵  and the Low Back Outcome Score,⁴⁶  do contain several questions related to the participation domain and may be considered depending on the region of the patient's injury.

Consider, for example, the care of a patient with an ankle sprain. The FAAM may be the PROM that a clinician identifies for use based on the fit of the instrument to the region of interest, instrument essentials, and patient friendliness. However, one consideration is that the FAAM is largely focused on functional ability.³¹  If the AT is approaching care from a patient-centered, whole-person perspective, coupling the FAAM with additional PROMs may be necessary, as the FAAM may evaluate only a limited scope of the HRQOL dimensions affected by an injury. A generic instrument (eg, Pediatric Quality of Life, DPA) could be considered depending on the HRQOL dimensions most relevant to the patient. In a recent case report, Fraser and Hertel¹⁵⁰  described the effect of a lateral ankle sprain on comprehensive function, HRQOL, and kinesiophobia. Through their use of multiple PROMs (ie, Godin Leisure-Time Exercise questionnaire, FAAM, Identification of Functional Ankle Instability, Tampa Scale of Kinesiophobia, Patient-Reported Outcomes Measurement Information System, EuroQoL), the authors were able to capture postinjury deficits across multiple ICF health domains and HRQOL dimensions and comprehensively manage the rehabilitation process while considering the patient's perspective via PROMs. In this case report,¹⁵⁰  they used many instruments, which may be unrealistic for everyday patient care. However, the diversity of these instruments provided the clinicians with unique information that helped drive treatment and emphasized patient-centered care. This example shows that whereas most of the PROMs are valid for use in managing patient care, patient goals and case details (eg, ICF health domains, health dimensions) may be strong factors influencing instrument selection.¹⁹ 

Of the PROMs reviewed, the MFA and Short MFA appeared to possess challenges for use in some patient groups related to athletic training. These instruments were originally developed^79,81  to assess musculoskeletal disorders in patients within the community and academic settings, which may limit their appropriateness for the athletic population. Although these scales are attractive because of their applicability to patients with a wide variety of musculoskeletal conditions,^79,81  their long length, due to the inclusion of questions that pertain to functioning over the entire body, makes them less patient (eg, completion time, survey fatigue) and clinician (eg, time required to score) friendly. Further, specific questions may be considered unacceptable for some patients because of their sensitive nature (eg, items related to sexual functioning). Based on previous findings,¹⁵  the MFA and Short MFA were used by ATs, albeit at relatively low percentages (10.9% for the MFA and 13.5% for the Short MFA). In some athletic training facilities, the MFA and Short MFA may be appropriate because of factors such as patient demographics (eg, age) or greater ability and organizational support to complete longer PROMs, although an instrument that is more focused on the body region of interest may provide a similar assessment in a shorter amount of time. Nonetheless, ATs interested in using the MFA or Short MFA should be aware of these factors when deciding if these instruments are appropriate and if they fit the intended use in the target patient population.

Time is the primary factor that influences whether clinicians implement PROMs.^14,15  In addition, the reality is that many ATs are relatively new to the assessment of patient outcomes as a routine component of care.^14,15  One type of PROM that may be appealing for clinicians who are beginning to implement outcomes assessment into care is the single-item instrument. Single-item instruments are arguably the easiest to incorporate into patient care because they consist of 1 question and take little time to administer, complete, and score. As a result, these measures are very patient and clinician friendly. Although single-item PROMs provide a quick glimpse of health in a short amount of time, these measures have limitations. Single-item PROMs do not provide as much information related to any specific ICF health domain or HRQOL dimension as multi-item measures. Also, the wording of some single-item instruments is vague and refers to health status or injury or illness in general and not specifically to characteristics such as pain or function. Thus, the exact aspect of health that the patient is reflecting upon when completing some single-item PROMs is unknown.¹⁰⁰  For example, the GROC instrument asks patients to compare their health between one point in time and another.¹⁰⁰  Not only does the GROC question ask patients to calculate a difference between time points, the health construct focused on by the patient could be related to a number of different aspects of health (eg, pain, function, ability to complete roles).¹⁰⁰  In addition, clinicians should consider whether an instrument has been modified from its original version and if those modifications are psychometrically sound. The GROC was originally developed as a 15-point scale;¹⁰¹  however, modifications (eg, 11-point,¹⁵¹  9-point,¹³³  7-point,¹⁴⁴  and 5-point¹⁵²  scales) are available. The abbreviated versions are frequently incorporated into patient care and used in research studies, yet limited information exists regarding their development. Athletic trainers should consider the use of single-item PROMs in practice because of their versatility and ease of use, and these instruments are helpful when an AT is starting to include outcomes assessment as a routine part of patient care. However, ATs should also be aware that the brevity of these instruments results in less information gained about the patient and the health condition than from multi-item measures.

A limitation of our report is that not all of the instruments used in athletic health care were included in this review. In an effort to report on instruments used in athletic training, investigators¹⁵  conducted survey research to identify the PROMs used most often by ATs. Although we believe that the PROMs reviewed in this report have the potential to support athletic health care and are commonly used in practice, there are likely other instruments with which ATs should be familiar. Furthermore, the landscape of PROM use in athletic training is ever changing. Thus, newer instruments developed in recent years may not have been reviewed in this report. Despite these limitations, we believe that our concise summary of PROMs used by ATs is a helpful resource for the profession as a whole, given ATs' relatively low use of and general lack of knowledge regarding PROMs, which appear to hinder the overall use of PROMs in athletic health care.

In general, the PROMs included in this report possess established and appropriate instrument essentials and clinical utility, supporting their use in patient care. With respect to the ICF health domains and HRQOL dimensions of health, the included PROMs generally focus on body structure and function as well as the physical functioning of the patient. Although that focus is not surprising and is typically helpful in caring for athletes, a sole focus on these components does not comprehensively capture the patient from a whole-person perspective. Thus, ATs with the primary goal of evaluating each patient as a whole person to support patient-centered care should consider a collection of PROMs as opposed to a single instrument. Efforts to make the use of PROMs in athletic training more routine will likely result in the development of new PROMs that are designed specifically for highly functional patients. As a result, in addition to considering the PROMs included in this report, ATs should stay abreast of newly developed PROMs.