Injury-related Psychological Distress and the Association with Perceived Running Ability in Injured Runners

Abstract

Objective:

Examine injury-related psychological distress and association with perceived running ability in injured runners.

Design:

Prospective longitudinal study.

Participants:

Forty-three patients with a running-related injury.

Main Outcome Measures:

Data collection at initial physical therapy visit and 12 to 16 weeks later included Optimal Screening for Prediction of Referral and Outcome - Yellow Flag (OSPRO-YF) and Athlete Fear Avoidance Questionnaire (AFAQ) for injury-related psychological distress, and University of Wisconsin Running and Recovery Index (UWRI) for perceived running ability. OSPRO-YF composite score, total yellow flags, and yellow flags in each domain (negative mood, fear-avoidance, positive affect/coping) were calculated.

Results:

UWRI score and OSPRO-YF composite score and yellow flags significantly improved over time, while AFAQ score and yellow flags in OSPRO-YF negative mood domain did not. AFAQ scores were significantly correlated with UWRI score at baseline, follow-up and change over time, while OSPRO-YF composite score and yellow flags were not. Baseline OSPRO-YF composite score and AFAQ score were not correlated with follow-up UWRI score.

Conclusions:

Injury-related psychological distress is elevated when injured runners start rehabilitation, and generally improves; however, negative mood and athletic fear-avoidance may persist. Higher athletic fear-avoidance is associated with lower perceived running ability at the same time point or interval.

Introduction

Running is a popular form of exercise as demonstrated by the 17.6 million runners registered for road races in the United States in 2019 (1). Running has both physical benefits, and mental benefits, such as lowering depressive symptoms and enhancing emotional well-being (2–4). Unfortunately, musculoskeletal injuries are common in running with an estimated 17.8 injuries in novice runners and 7.7 injuries in recreational runners per 1000 hours of running (5). Running injuries can result in missed training sessions, altered training schedules, missed competitive events, and visits to medical professionals (6, 7), which in turn may increase the likelihood of psychological distress (8).

Rehabilitation for running injuries has historically been grounded in a biomedical approach that focuses on addressing tissue pathoanatomic factors and abnormal running biomechanics (9). However, pain and running limitations can persist in some runners even after tissue pathology and running biomechanics have been properly identified and treated (10). The phenomenon of persistent pain and disability after a musculoskeletal injury has been described in the fear-avoidance model of musculoskeletal pain (11). Per the model, if pain from injury is perceived as a threat, the patient will catastrophize the pain, develop a fear of movement and re-injury that leads to activity avoidance, which in turn results in disuse, disability and depression (11). Fear-avoidance beliefs have been shown to be elevated and associated with physical function in non-athletic and athletic patients with musculoskeletal injury (12, 13). To date, the fear-avoidance model has not been examined in an injured runner population. It is unknown if injured runners have heightened fear-avoidance since only 6% of running injuries result in complete cessation of running (14).

Pain Catastrophizing Scale (15), Tampa Scale for Kinesiophobia (16), and Fear-Avoidance Beliefs Questionnaire (17) have been used to measure psychological constructs in the fear-avoidance model. Administering multiple questionnaires can be useful to gain a more comprehensive understanding of the psychological status of the patient, but it is also a time burden to the patient and clinician. The Optimal Screening for Prediction of Referral and Outcome - Yellow Flag (OSPRO-YF) is a concise tool that screens for injury-related psychological distress in three domains (negative mood, fear-avoidance, and positive affect/coping) (18) and has been validated in a general population of patients with musculoskeletal injury (19, 20). It is possible that athletes experience fear-avoidance related to sport participation that is not captured in a general screening tool. For this reason, the Athlete Fear Avoidance Questionnaire (AFAQ) was created (21). The AFAQ measures emotions and beliefs that are specific to injured athletes such as their feelings regarding how their injury will affect their athletic performance and beliefs about return to sport. The OSPRO-YF focuses on general physical activity, emotions, and beliefs about pain and living with pain. Neither OSPRO-YF nor AFAQ have been studied in an injured runner population.

It has been recommended that physical therapists use a psychologically-informed approach to treat musculoskeletal pain conditions, which requires identifying patients with elevated psychological distress and providing alternative treatment methods, if needed (22). This recommendation has been incorporated into recent clinic practice guidelines for knee and hip pain (23–25). Obtaining a better understanding of injury-related psychological distress in injured runners can help discern if injured runners require this modified treatment approach to facilitate recovery. The purpose of this study was to examine injury-related psychological distress in injured runners and the association with perceived running ability 12 to 16 weeks later. We hypothesized that injury-related psychological distress would be elevated in injured runners at the initiation of rehabilitation, improve over time, and show a negative association with running perceived ability.

MATERIALS AND METHODS

Study Overview

This study was a secondary, exploratory component of a multi-center, prospective, longitudinal study evaluating psychometric properties of the University of Wisconsin Running and Recovery Index (UWRI) (26). Demographic information and UWRI questionnaire responses were collected at all sites at the initial physical therapy visit (baseline) and 12 to16 weeks later (follow-up). The follow-up time frame was chosen to address the main study aim of examining psychometric properties of UWRI and was selected as it is reasonable to expect clinically meaningful change to occur, while minimizing participant loss associated with longer time frames. Participants who enrolled at TRIA Orthopedic Center, Bloomington, MN completed injury-related psychological distress questionnaires at both time points as an exploratory analysis of the primary multi-center study Demographic information and questionnaire responses were collected with Research Electronic Data Capture (REDCap).

Subjects

Patients with a running-related lumbopelvic or lower extremity injury who presented to a running injury rehabilitation program at TRIA Orthopedic Center were eligible to participate in the study. Additional inclusion criteria included age 14 years or older, English-speaking, and no surgery at the injury site in the past 12 months. There were no inclusion criteria for baseline mileage or years of running experience due to the inclusive nature of the primary study. A total of 72 participants were enrolled in the study and gave electronic informed consent (adults) or assent with parental consent (minors). Of these, 29 participants were excluded because they did not complete questionnaires; thus, 43 participants comprised the study sample (Figure 1). This study was approved by the institutional review board at University of Wisconsin-Madison.

Figure 1.

Study flow diagram.

Demographic Information

Participant characteristics collected by self-report included age, sex, body mass index (BMI), previous injury (yes/no), and self-reported weekly running volume (Table 1). Additionally, participants self-selected their running status, defined as novice (new to running), recreational (regular running with or without race participation for enjoyment or health and wellness purposes), or competitive (purposeful training to improve performance without financial gain). The duration of the current injury and its location, with confirmation by one of five treating physical therapists, were also recorded.

Table 1.

Demographic information for the study sample and participants excluded from analysis. Continuous variables are mean (SD).

Variable		Study Sample (n = 43)	Excluded (n = 29)
Age (years)		36.5 (13.0)	35.8 (12.1)
Sex (n)	Females	31	16
	Males	12	13
BMI (kg/m2)		23.9 (3.3)	23.2 (4.0)
Baseline to follow-up testing (weeks)		15 (2.5)
Prior running-related injury (n)		19	9*
Injury duration (months)		3.4 (1.5)	3.4 (1.7)
Injury location (n)†	Low back	3	0
	Pelvis	2	1
	Hip	6	3
	Thigh	2	1
	Knee	15	14
	Calf/Shin	8	5
	Ankle	7	6
	Foot	8	4
Running status (n)	Novice	1	0
	Recreational	28	22
	Competitive	14	6
Running experience (years)		14.4 (10.7)	10.5 (8.1)
Weekly running volume (km)		42.0 (23.1)	37.2 (22.5)

^*Prior injury data missing for 2 excluded participants

^†7 participants in the study sample and 4 excluded participants had injuries in multiple locations

Questionnaires

University of Wisconsin Running and Recovery Index (UWRI)

The UWRI is a 9-item evaluative, patient-reported outcome measure assessing perceived running ability following injury (27). An example question is “How has your weekly mileage or weekly running time changed as a result of your injury?” Each item is scored 0 to 4 on a Likert scale, with responses varying according to the question. The total score ranges from 0 to 36 points. Higher scores mean better perceived running ability, and 36 points means a return to preinjury running ability. The UWRI has been shown to have acceptable internal consistency (α= 0.82), excellent test-retest reliability (ICC = 0.93), and construct-related validity. (27). In addition, the UWRI is responsive to change, and improvement of 5 points represents minimal important change and improvement of 8 points represents minimal clinically important difference (26).

Optimal Screening for Prediction of Referral and Outcome - Yellow Flag (OSPRO-YF)

OSPRO-YF is a multidimensional tool for assessing injury-related psychological distress in a general population (18). The original version includes 17 items obtained from 11 validated psychological questionnaires that encompass 3 domains (negative mood, fear-avoidance, and positive affect/coping). Items are summed, with reverse scoring for pain resilience items, to create a composite score that ranges from 6 points to 89 points (18). Higher scores indicate higher levels of injury-related psychological distress. In addition, the presence of yellow flags, indicating elevated psychological distress, can be determined by scores in the top quartile (elevated vulnerability) for negative mood and fear-avoidance domains and from scores in the bottom quartile for positive affect/coping domain (decreased resiliency) (18). The total number of yellow flags on the entire tool ranges from 0 to 11; the number of yellow flags possible in each domain is 3 for negative mood, 5 for fear-avoidance, and 3 for positive affect/coping. OSPRO-YF has good concurrent validity and has shown some ability to predict 12-month outcomes for common musculoskeletal conditions in a general population (18, 20, 28).

Athletic Fear Avoidance Questionnaire (AFAQ)

AFAQ is a 10-item questionnaire that measures sport injury-related fear-avoidance (21). An example item is “I believe that my current injury has jeopardized my future athletic abilities.” Each item is rated 1 (not at all) to 5 (completely agree) with a total score ranging from 10 to 50 points. Higher scores represent greater fear-avoidance related to sports. The AFAQ was validated on university athletes from various sports including soccer, rugby, football, basketball and hockey (21). The AFAQ demonstrates good internal consistency (Cronbach α = 0.81) and concurrent validity with the Fear-Avoidance Beliefs Questionnaire – Physical Activity (r=0.35) and the Pain Catastrophizing Scale (r = 0.587) (21).

Statistical Analysis

Sample size estimation was based on correlations of OSPRO-YF and AFAQ scores with patient-reported measures of physical function reported in other studies (r = 0.57, r = 0.32; respectively) (13, 19). An online calculator (https://www.ai-therapy.com/psychology-statistics/sample-size-calculator) determined that a sample of 27 to 50 runners was needed to detect correlations between 0.32 and 0.57 with 80% power and significance level of α = 0.05.

Statistical analysis was performed with SPSS Statistics Version 26 (IBM Corporation, Armonk, NY). Descriptive statistics were generated for demographic variables and questionnaire scores. Demographic variables were compared between the study sample and participants who were excluded because of incomplete questionnaire data using independent samples t-test or Chi Square test, as appropriate.

UWRI, OSPRO-YF composite and AFAQ scores at baseline and follow-up were compared with paired t-tests to evaluate change over time. The distribution of OSPRO-YF yellow flags as well as yellow flags on the domains of OSPRO-YF were compared between time points using the Wilcoxon signed rank test.

Bivariate associations of OSPRO-YF and AFAQ scores with UWRI score were examined at baseline, follow-up, and in the change across 12–16 weeks (follow-up value – baseline value). In addition, the association of OSPRO-YF and AFAQ scores at baseline with UWRI score at follow-up was examined. Examining associations at multiple time points and time intervals can provide insight into patterns of association that can inform clinical application. Associations of OSPRO-YF composite and AFAQ scores with UWRI score were determined with Pearson correlation, while associations of OSPRO-YF yellow flags with UWRI score were examined with Spearman rank correlation. Bonferroni correction was performed to adjust for the 9 tests of correlation (p=0.05/9=0.006). Statistical significance was set at p<0.05 for the other analyses.

RESULTS

Study Sample

Demographic variables for participants in the study sample and those excluded from the study did not differ (Table 1). The study sample was mainly comprised of females (72%) and recreational runners (65%). At baseline, 13 participants (30%) were unable to run and 30 participants (70%) had reduced running mileage due to their injury.

Change in Questionnaire Scores over Time

Scores on each questionnaire are found in Table 2. UWRI score and OSPRO-YF composite score significantly improved from baseline to follow-up (p<0.001). The mean magnitude of improvement in UWRI score was 9.1 points, which surpasses the threshold for minimal clinical important difference (26). AFAQ score did not significantly change (p = 0.094). Further examination of the AFAQ data showed that 14 participants (33% of the sample) had an increase of at least 1 point (range: 1 to 15 points) in AFAQ score from baseline to follow-up. Of the 14 participants who had an increase in AFAQ score, 5 participants had a decrease in UWRI score.

Table 2.

Mean (SD) scores on each questionnaire at baseline and follow-up testing.

Questionnaire	Baseline	Follow-up	Change
UWRI (points)	14.9 (5.5) range: 4 to 25	24.1 (7.0) range: 7 to 36	9.1 (7.3)* range: −7 to 28
AFAQ (points)	21.8 (8.1) range 10 to 39	19.7 (7.7) range: 10 to 41	−2.0 (7.6) range: −23 to 15
OSPRO-YF composite (points)	31.9 (9.5) range: 15 to 49	27.6 (8.7) range: 13 to 54	−1.2 (2.4)* range: −16 to 10
OSPRO-YF yellow flags (n)	3.0 (3.0) range: 0 to 9	1.8 (2.4) range: 0 to 9	−1.2 (2.4)* range: −7 to 4

^*Significant change from baseline to follow-up, p<0.05.

The number of yellow flags in each domain of the OSPRO-YF are found in Table 3. The distribution of the number of OSPRO-YF yellow flags significantly changed from baseline to follow-up (p=0.003). More specifically, fear-avoidance and positive coping/affect domains showed a greater number of patients exhibiting no yellow flags at follow-up. No significant change was found in the distribution of yellow flags from baseline to follow-up for the negative mood domain (p=0.285). Examination of the data showed that when compared to baseline, there were 3 more participants who exhibited no yellow flags and 3 more participants who exhibited 2 yellow flags at follow-up.

Table 3.

Number of yellow flags on each domain of OSPRO-YF.

Domain	Time Point	Yellow Flags
		0	1	2	3	4	5
Negative Mood	Baseline	25	11	4	3
	Follow-up	28	7	7	1
Fear-Avoidance*	Baseline	22	5	3	3	9	1
	Follow-up	29	6	3	2	3	0
Positive Affect/Coping*	Baseline	21	9	9	4
	Follow-up	27	10	5	1

^*Significant change from baseline to follow-up, p<0.05.

Correlations

At baseline, AFAQ score (r= −0.432, 95% CI [−0.648, −0.152], p = 0.004) was significantly negatively correlated with UWRI score (Figure 2), while OSPRO-YF yellow flags and OSPRO-YF composite score were not (p = 0.023 and p = 0.093 respectively).

Figure 2.

Correlations of AFAQ score at baseline with UWRI score at baseline.

At follow-up, AFAQ score (r= −0.672, 95% CI [−0.809, −0.466], p<0.001) was significantly negatively correlated with UWRI score (Figure 3). OSPRO-YF yellow flags and OSPRO-YF composite score were not (p=0.074 and p=0.048, respectively).

Figure 3.

Correlations of AFAQ score at follow-up with UWRI score at follow-up.

Change over time in AFAQ score (r = −0.606, 95% CI [−0.766, −0.374], p<0.001) was significantly negatively correlated with change over time in UWRI score but OSPRO-YF composite score and OSPRO-YF yellow flags were not (p=0.038 and p=0.013, respectively) (Figure 4).

Figure 4.

Correlation of AFAQ change score with UWRI change score.

Baseline values for OSPRO-YF composite score, OSPRO-YF yellow flags and AFAQ score showed no significant correlation with UWRI score at follow-up (p=0.639, 0.613, 0.370; respectively).

DISCUSSION

This study examined injury-related psychological distress and the association with perceived running ability in runners receiving physical therapy for a musculoskeletal injury. Our hypotheses that injury-related psychological distress would be elevated in injured runners at initiation of rehabilitation, improve over time and show association with perceived running ability were generally supported. OSPRO-YF composite score and yellow flags significantly improved over the follow-up period; however, yellow flags on OSPRO-YF negative mood domain and AFAQ score did not. AFAQ and UWRI scores were negatively correlated at baseline, follow-up and for the change between time points, while OSPRO-YF composite score and yellow flags were not, indicating that higher fear-avoidance is associated with lower perceived running ability. On the other hand, OSPRO-YF and AFAQ scores at baseline were not correlated with UWRI score at follow-up. In summary, injured runners demonstrate elevated psychological distress at the start of rehabilitation. Psychological distress generally improves over time, although negative mood and athletic fear-avoidance can persist. Additionally, higher athletic fear-avoidance is associated with a perception of lower running ability when assessed together at the same time point or interval.

OSPRO-YF and AFAQ assess different aspects of injury-related psychological distress and generated different results in injured runners. OSPRO-YF composite score and the number of yellow flags significantly decreased over time. It is possible that the change over time indicates injury-related psychological distress was elevated at the initiation of rehabilitation, similar to a general musculoskeletal injury population (12, 19). In fact, the OSPRO-YF composite score at baseline and the magnitude of improvement are similar to findings in a general musculoskeletal injury population that had a 12-month follow-up period, as opposed to our 12–16 week follow-up (28). These findings add to the literature by demonstrating the presence of psychological distress in injured runners when they begin rehabilitation.

Yellow flags on OSPRO-YF negative mood domain and AFAQ score did not improve from baseline to follow-up. Previous research has established that negative mood and fear-avoidance can increase around the time of return to sport in athletes (29–32). Negative mood could also persist in runners due to prolonged rehabilitation or interruption of training (14, 33). Additionally, runners that sustain injuries have higher negative affect prior to injury compared to uninjured runners (34). This pre-injury distress may result in perpetuation of negative mood during rehabilitation as this may be a trait of some runners rather than a temporary state. Although the participants in this study demonstrated a significant and clinically meaningful improvement in perceived running ability, they generally did not achieve pre-injury perceived running status based on UWRI score, indicating continued limited training. Reduced training volume and intensity have been associated with a more negative mood (35). Persistent negative mood exists in some injured runners with reduced training volume and intensity, who rely on running as a primary coping mechanism for non-injury related psychological distress, such as those who experience exercise addiction (36).

The lack of change in AFAQ score could be explained by two phenomena. First, the mean baseline AFAQ score approximates the lowest score (low fear-avoidance) for the questionnaire, leaving little room for improvement at follow up. In fact, the mean baseline score of 21.8 points is even lower than other injured athlete populations where mean AFAQ scores were reported as 26, 23.7, and 22.1 points respectively (13, 21, 37). Secondly, one third of the sample showed an increase in AFAQ score at follow-up, demonstrating the variability of psychological response. A clinically significant change in fear-avoidance is unknown, and we observed that not all runners with an increase in AFAQ reported a decline in perceived running ability. It is possible that a slight increase in fear-avoidance does not interfere with running ability since many runners continue to run despite injury (14). The variability in psychological response suggests that athletic fear-avoidance may both influence perceived running ability and be influenced by perceived running ability. For example, high fear-avoidance early in rehabilitation might limit advancement of a runners training volume and intensity. Conversely, advancement in training volume and intensity could trigger athletic fear-avoidance. Here a high training volume near the volume they were running at prior to injury may be reached, and the potential perceived threat of higher training volume may trigger an increase in athletic fear-avoidance.

Further research is needed to determine the threshold where alternative treatment approaches should be considered when there is an increase in fear-avoidance and a decline in perceived running ability. This is important as fear-avoidance is the most important factor associated with decreased physical function in athletes (13). Fear-avoidance is modifiable with interventions such as cognitive behavioral therapy (38, 39), exposure therapy (38), and mindfulness training (40). An additional variable that may influence persistent negative mood and athletic fear-avoidance in runners undergoing physical therapy may be the confidence the runner has in their health care providers’ skill, judgement, and recommendations, otherwise known as therapeutic alliance. A strong therapeutic relationship may allow an injured runner to progress with less fear that they may be doing harm as they increase activity. Treatment targeted at building a strong therapeutic alliance may improve outcomes (41).

The examination of associations between injury-related psychological distress measures and perceived running ability revealed an interesting pattern. Although we found association between AFAQ and UWRI scores, no associations were found between OSPRO-YF and UWRI scores. The statistically significant correlation of the AFAQ was likely because athletic activity is the focus of both AFAQ and UWRI questionnaires, while the OSPRO-YF was designed for a general musculoskeletal injury population. The finding highlights the importance of using an athletic-oriented fear-avoidance measure in athletic populations. Additionally, it is important to note that despite the lack of association with UWRI, the OSPRO-YF can be used with runners as a screening tool to guide referral for psychologically informed intervention as it screens psychological domains beyond athletic fear-avoidance such as negative mood and positive affect/coping.

This study found no association between baseline injury-related psychological distress and perceived running ability 12–16 weeks later. Other researchers have similarly failed to find an association between baseline psychological factors and function at a later time point after musculoskeletal injury (28, 42). However, associations between AFAQ change scores provide preliminary support for serial monitoring of injury-related psychological distress in runners. Serial monitoring may identify situations when a patient’s psychological response to injury is misaligned with the physical recovery. Also, prolonged increase in psychological distress may be associated with persistent symptoms (43, 44). Thus, it is important to establish a baseline to identify the trajectory of injury-related psychological distress over time. We are careful not to imply causation from this correlational analysis, and larger-scale studies will be necessary to identify if injury-related psychological factors predict running ability or other functional outcome.

The main strength of this study is that it establishes an initial framework for understanding injury-related psychological distress in an injured runner population. This includes using a measure (OSPRO-YF) not previously studied in an athletic population. The results of this study add to the growing evidence that injury-related psychological distress, specifically fear-avoidance, is elevated after musculoskeletal injury and is associated with perceived function. There currently is no consensus on the best way to monitor injury-related psychological distress, and the results of this study highlight the considerations of using a general versus sports-oriented tool. A limitation of this study is the relatively small sample size that precludes multi-factorial modeling with variables such as the number of rehabilitation visits, injury chronicity, current running status, history of previous injury, and onset of new injuries. Another limitation is the UWRI includes questions related to frustration and confidence that may have influenced the magnitude of correlation with the injury-related psychological distress questionnaires. The UWRI is a patient reported outcome measure and therefore measures only perceived running ability and not training volume or actual running ability which is a limitation of the study. In addition, there was a greater proportion of females in the study that could make the findings less generalizable to injured male runners. Rehabilitation was not controlled in this study; thus, it is not possible to make inferences about study results based on the type of rehabilitation provided. Finally, the relatively short follow-up may be insufficient for full functional and psychological recovery (45).

In conclusion, injured runners may demonstrate elevated injury-related psychological distress at the initiation of rehabilitation. Injury-related psychological distress generally improves over time, except negative mood and athletic fear-avoidance. Injury-related psychological distress, specifically athletic fear-avoidance and perceived running ability are negatively associated when these measures are assessed at the same time points or in the same time interval. Study results provide preliminary support for using a psychologically-informed approach to treating musculoskeletal injury in runners. Further research is needed to determine if OSPRO-YF and AFAQ are optimal measures of injury-related psychological distress in runners and what treatment methods should be used for those with elevated distress. In addition, future research could examine if elevated injury-related psychological distress contributes to future injury and persistent running limitations (34, 46).

Highlights.

• Injured runners exhibit psychological distress at the start of rehabilitation

• Negative mood and athletic fear-avoidance beliefs may persist in some injured runners

• Higher athletic fear-avoidance is associated with lower perceived running ability