Many psychological factors influence an injured athlete's perception of symptom severity (Domenech et al., 2014), ability to rehabilitate from injury (Levy et al., 2008), and ability to return to sport (Ardern et al., 2012). Psychological factors such a high optimism (Brewer et al., 2006; Thomee et al., 2006), self-motivation (Brewer et al., 2003), athletic self-identity (Brewer and Cornelius, 2010), and perceived social support (Brewer et al., 2003; Brewer et al., 2000) positively impact injury rehabilitation and return to sport. Alternatively, psychological factors such as high levels of kinesiophobia (Domenech et al., 2013; Kvist et al., 2005), pain catastrophizing (Domenech et al., 2014), and depressive symptoms (Galambos et al., 2005) can negatively impact sports injury treatment outcomes.
With the growing body of evidence linking psychological measures to sport injury treatment outcomes, there is a need for sports medicine clinicians and researchers to familiarize themselves with psychological assessments used in the context of sports medicine injury treatment. Though the concepts are often intuitive, sports medicine providers may be unfamiliar with assessment tools used to measure psychological factors. This lack of familiarity makes it difficult to critically assess new literature in this area or to employ the assessment tools in practice. Self-report measures are advantageous in that they are simple to administer and are the most common method used to assess psychological factors in sports medicine literature. However, it is first essential to confirm that the tool has adequate internal consistency (a measure of internal reliability) prior to performing further validation studies (John and Soto, 2007) or utilizing the tool in a clinical context.
The primary aim of this review is to identify all self-report psychological assessment tools and specific psychological traits investigated in sports-related musculoskeletal injury or concussion treatment outcomes studies reported in the English literature. A secondary aim is to identify the proportion of these assessment tools with adequate internal consistency, which can be utilized as a starting point for evaluation since it is a single, consistently reported reliability metric.
A systematic search was performed with reporting of results per PRISMA guidelines (Figure 1)(Moher et al., 2009). A search of relevant studies published from 1950 (earliest indexed article that met our search terms) to November 1st, 2019 was initially performed on PubMed. Initial PubMed MESH terms used were ([psychological outcomes] OR [psychological measures]) AND ([sports medicine] OR [sports injury] OR [ACL] OR [knee] OR [ankle] OR [shoulder] OR [back] OR [hip] OR [elbow] OR [concussion]) which yielded 2759 human studies reported in English. Searches were then performed in Google Scholar, Scopus, and SPORTDiscus as well as review of citations used in relevant studies to identify an additional 119 studies (n = 878 total). The searches were performed independently by two authors with review of all discrepant selections by a third author. Studies were reviewed for predetermined inclusion criteria, including report of a sports injury treatment outcomes study, use of a self-report psychological assessment tool before and/or after injury treatment, and a requirement of publication in English in a peer-reviewed journal (Table 1). The database searches yielded a total of 2878 studies. A total of 2612 studies were excluded due to clear failure to meet 1 or more inclusion criteria based on the content of the study title or abstract alone. The entire manuscripts of the remaining 266 studies were reviewed, resulting in exclusion of a further 114 studies. This yielded a total of 152 sports-related injury out a pre-treatment or post-treatment, self-report psychological assessment tool. For this review, a self-report assessment tool was defined as one available in paper or electronic format that is completed by the patient with or without supervision of clinical staff. None of the included psychological assessment tools required completion by an individual other than the patient. Scoring could occur via manual scoring (scoring by hand) or with the use of automated scoring tools.
A total of 34 unique self-report psychological tools were identified that were employed either pre-intervention or post-intervention in a musculoskeletal sports injury or concussion treatment study. After identifying all self-report psychological assessment tools, citations for the original publication describing the assessment tool, and, as available, the initial validation studies, were compiled.
The validity of a scale is limited by its reliability (John and Soto, 2007); thus, it is essential to determine that a scale is reliable before determining validity. A key measure of reliability for self-report measures is internal consistency; often reported as Cronbach's alpha, this is a measure of the correlation between different items within the same assessment tool. A higher value implies that individual items in a scale or subscale are measuring the same psychological trait. The recommended minimum value for Cronbach's alpha is 0.70 (Bland and Altman, 1997; DeVellis, 2016; Nunnally and Bernstein, 1994), and the number of assessment tools with internal consistency (both in whole and all subscales) above 0.70 and 0.80 were determined.
Results were organized by 1) psychological traits that generally have a "positive" influence on treatment outcomes into categories based on the factors or traits being assessed including personality traits (n = 2 tools); self-motivation (n = 3); coping strategies (n = 2); perceived social support (n = 2); athletic self-identity (n = 1); and optimism and self-efficacy (n = 5), 2) traits that have "negative" influence on outcomes including fear-avoidance of pain or movement (n = 5 tools); psychological distress (n = 5); and depressed mood (n = 8), as well as 3) general measures of mental-health related quality of life (n = 2). Finally, the most commonly utilized self-report psychological assessment tools, defined as those that were utilized in more than 5 sports medicine injury outcomes studies, were identified and summarized.
A total of 34 distinct self-report psychological assessment tools used across 152 studies were identified within the sports medicine literature pertaining to treatment of concussions or injuries of the shoulder, elbow, knee, hip, ankle, or foot. The median number of items per scale was 20 (range: 4 to 567, lengths of each individual assessment tool are listed in Table 2). Assessments of psychological traits that generally have a "positive" influence on treatment outcomes (n =15 tools) included personality traits (n =2 tools); self-motivation (n = 3); coping strategies (n = 2); perceived social support (n = 2); athletic self-identity (n = 1); and optimism and self-efficacy (n = 5). Assessments of psychological traits that have a "negative" influence on outcomes (n = 17 tools) including fear-avoidance of pain or movement (n = 4 tools). Finally, two self-report psychological assessment tools for general mental-health related quality of life were identified.
Internal consistency was reported for the whole scale and all subscales of 31 of 34 (91%) assessment tools.
The three assessment tools without demonstrated internal consistency of all or part of the assessment tool are the Psychovitality Scale (no data reported) (Gobbi and Francisco, 2006), the Emotional Responses of Athletes to Injury Questionnaire (ERAIQ) (no data reported)(Smith et al., 1990), and the ACL deficiency Quality of Life scale (ACL-QoL) (no data reported for subscales; total scale internal consistency reported as [alpha] = 0.93)(Mohtadi, 1998). Of the 34 assessment tools, 28 (82%) met the cutoff of [alpha] [greater than or equal to] 0.70 for adequate internal consistency of both the whole scale and each subscale (12 tools (35%) had an [alpha] between 0.70-0.79 and 16 tools (47%) had an [alpha] [greater than or equal to] 0.80). Three assessment tools with adequate reporting of internal consistency failed to meet a minimum internal consistency of [alpha] = 0.70 for the total scale or one or more subscales; these included the Sports Injury Rehabilitation Beliefs Survey (SIRBS) severity subscale ([alpha] = 0.52) (Taylor and May, 1996), the Illness Perception Questionnaire-Revised (IPQ-R) (subscales [alpha] = 0.67-0.89) (Moss-Morris et al., 2002), and the Swedish universities Scales of Personality (SSP) social desirability subscale ([alpha] = 0.59) (Gustavsson et al., 2000).
Six psychological assessment tools have been utilized in 5 or more sports injury treatment outcome studies: the Pain Catastrophizing Scale (PCS) (Sullivan et al., 1995), Tampa Scale for Kinesiophobia (TSK) (Huang et al., 2019; Kori et al., 1990) and TSK-11 (Woby et al., 2005)), Sports Injury Rehabilitation Beliefs Survey (SIRBS) (Taylor and May, 1996), Beck Depression Inventory Fast Screen (BDI-FS) (Steer et al., 1999), Emotional Responses of Athletes to Injury Questionnaire (ERAIQ) (Smith et al., 1990), Short Form 36 Questionnaire Mental Health Component (SF-36 MCS) (Ware Jr and Sherbourne, 1992).
The link between personality measures and sports injury treatment outcomes is not commonly investigated (Table 2). Two scales of personality trait, The Minnesota Multi-phasic Personality Inventory (MMPI) (Hathaway and McKinley, 1940) (since revised to MMPI-2 and MMPI-2-RF)(Hathaway et al., 1989) and the Swedish Universities Scale of Personality (SSP) (Gustavsson et al., 2000) have been utilized in a sports medicine context (Swirtun and Renstrom, 2008; Wise et al., 1979), with internal consistency of subscales ranging from 0.59-0.92 and length ranging from 91-567 items.
Self-motivation is the ability to initiate activity without the influence of others (Dishman and...