Methods

Subject

This study included varsity University of Calgary wrestlers, football and ice hockey players of both sexes. Potential subjects were identified by determining which teams were given the SCAT as part of their preseason medical evaluation. Approval was obtained from the Health Ethics Research Board at the University of Calgary. Consent was ascertained at the time of each athlete’s yearly physical. All collision athletes undergo a preseason medical by a physician at the university medical clinic. This includes a detailed questionnaire concerning past medical history, with specific questions regarding concussion history. Any subsequent medical concerns are dealt with by the team athletic therapist or team physician at the university medical clinic. Electronic medical records (EMR) and the preseason questionnaire were used to determine previous concussion history. A participant was considered “previously concussed” (PC) if they self-reported ever having a previous concussion on their preseason medical questionnaire or if a concussion was diagnosed by the team doctor during a playing season before the baseline SCAT. Inclusion criteria included: (1) individuals on the active roster for varsity wrestling, hockey or football; (2) consent form completion and (3) completion of a baseline SCAT. Exclusion criteria included individuals who were concussed at the time of SCAT administration; however, no individuals were excluded for this reason.

Procedure and statistical analysis

SCAT forms were collected over 3 years (2005–2007) and only an athlete’s first baseline SCAT was used in the analyses. Data were collected from the SCAT forms and compiled based on response frequency. Data were grouped according to: man, woman, NC and PC. The first year of data collection was done retrospectively, after the SCAT had already been administered to the subjects by team athletic therapists. In subsequent years, athletic therapists and team doctors who were aware of the goals and purposes of this study administered the SCAT.

Each symptom on the post-concussion symptom scale (PCSS) score was reviewed independently. If a symptom was left blank then a score of 0 was assumed. If more than one score was circled for a symptom then the higher score was used. To determine normative values for the PCSS, the mean, mode and median total symptom scores were calculated from the response frequencies for each athlete. Athletes with a PCSS score of 0 were determined and the mean total symptom score was calculated for each group excluding those subjects. Based on frequency counts for each symptom the mode and median for each symptom were also calculated. Finally, the top five symptoms with higher scores were determined for each group based on the symptoms with the highest percentage of responses scoring one or greater.

The cognitive assessment was analysed separately and each test within the cognitive assessment was treated independently. A subject’s SCAT was therefore included in the calculations even if one component was incomplete. The five-word recall was assessed based on the number of words correctly recalled. Immediate and delayed recall was assessed for each of the five groups independently; if a space was left blank then it was assumed that the word was not recalled. Normative values for five-word recall were determined by calculating the mode and median number of words remembered and the proportion of subjects for each score.

Months in reverse order was assessed on a pass/fail basis in which a subject passed if they were able to recite the 12 months in reverse order with no mistakes. The test was considered a fail if any months were in the wrong order or missed. The percentage of subjects who passed and failed for each of the five groups was calculated.

Digits backwards was assessed in a similar fashion to the five-word recall. The highest number of digits correctly recited was recorded. If spaces were blank it was assumed that the number of digits was not achieved. To determine normative values for digits backwards the mode and median number of digits recited was calculated for each group along with the proportion of athletes for each score.

Results

This study included 260 subjects aged 17–32 years, with a mean age of 20.5 years. Of the 260 subjects, 190 were men and 70 were women; 167 had no previously reported concussions (NC) and 93 reported being previously concussed (PC). In addition, 39.0% of men and 27.1% of women reported a previous concussion. In the first year of data (2005) 49 SCAT were collected, 119 in 2006 and 92 in 2007.

The PCSS score was completed by all 260 subjects. Of 260 subjects, 61 did not complete the “follow-up symptoms only” section (23.5%). All (100%) subjects either did not fill out or put a score of 0 for the symptom other. Women had the highest mean total PCSS score at 6.39, whereas men had the lowest at 3.52 (table 1). Women also had the lowest number of “total 0s” (table 1.).

Figures 1–3 show the PCSS score frequencies for all five groups. The highest total baseline PCSS score (49) was reported by a NC female subject (fig 2).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Frequency count of total post-concussion symptom score (PCSS) for all participants.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Frequency count of total post-concussion symptom score (PCSS) for men and women.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3

Frequency count of total post-concussion symptom score (PCSS) for never concussed and previously concussed participants.

The mode and median PCSS scores for individual symptoms for all subgroups was 0, with the exception of a median score of 1 for fatigue/low energy among female athletes. The five most frequently reported symptoms are reported in table 2. If more than one symptom had the same frequency, all symptoms were reported. (see Appendix for frequency counts of individual symptoms (table A1 and table A2)).

The five-word recall was completed by 249 subjects. All subject groups were able to recall at least four words on immediate recall (table 3). PC subjects had the highest frequency of recalling five words (98.9%), whereas NC subjects had the lowest (94.9%; table 3).

For delayed five-word recall (table 4), the majority of subjects were unable to recall five words, with the exception of the female group. Women had the highest frequency of recalling five words (60.0%), whereas men had the lowest (29.6%). NC had the highest number of subjects who could not recall any words at 6.4%. The median and mode immediate recall score was 5/5 for all groups. On delayed recall, women scored 5/5 on both median and mode, whereas all remaining groups scored a median and mode of 4 and 5, respectively.

Months in reverse order was completed by 249 subjects. Women had the highest pass percentage (96.7%), whereas NC had the lowest (90.4%; table 5).

Digits backwards was completed by 247 subjects. PC subjects had the highest frequency of scoring six digits (61%), whereas NC had the lowest (46%; table 6). Although some subjects were unable to get 0 (three digits) no subjects were unable to get 1 or 2 (four and five digits, respectively; table 6). All participants, women and PC scored a median of 6 on reverse digits, whereas men and NC scored a median of 5.

Discussion

The data collected in this study show that the mean total baseline PCSS score for varsity collision sport athletes ranged between 3.5 and 6.5. This is in keeping with other studies that have found mild baseline symptom scores.10

Women had the highest baseline total symptom score and were more likely to rate their symptoms greater than 0. This result is consistent with Covassin et al,10 who found that female athletes endorsed a significant number of mild baseline symptoms compared with male athletes. PC athletes also reported a higher total baseline symptom score then NC athletes, which mirrors the findings of Bruce and Echemendia.11

The most commonly reported symptom for all groups was fatigue or low energy. Drowsiness and neck pain also ranked in the top five for all athletes. Men, NC and PC had the same top symptoms in a different order; this is consistent with the findings reported by Covassin et al.10 The increased frequency of these symptoms could be explained by a number of factors. Fatigue or low energy and drowsiness are common symptoms in athletes, which may be a result of intense training.12 Varsity athletes at the University of Calgary are also required to be registered in at least three courses and maintain a minimum grade point average of 2.0, adding to their daily demands.13

Women scored higher in all baseline cognitive tests on the SCAT compared with men. This outcome was similar to the findings of Covassin et al,10 in which female athletes scored significantly better on verbal memory than male athletes. PC athletes scored better than NC athletes in all cognitive tests except delayed five-word recall. These results are the opposite of what was expected and could suggest that NC athletes may not try as hard in baseline testing. It could also be speculated that as a result of their previous concussion history PC athletes have more experience with cognitive testing and that the results found in this study represent a practice effect.

The section of the SCAT labelled “follow-up symptoms” was not completed by 23.5% of subjects, suggesting that the instructions given were unclear about the purpose of this section.

The subgrouping used in this study assumed that none of the athletes experienced an undiagnosed or unrecorded concussion. It was also assumed that the athletes were given their SCAT in appropriate locations, and that the tests were administered properly without any distraction or other extenuating factors. Finally, it was also assumed that the subjects exerted full and consistent effort on their SCAT evaluations.

As the first year of the study was done retrospectively it was limited by the fact that the authors were not present at the time of data collection (SCAT testing); however, this only represents 49 out of 260 SCAT. In subsequent years data were collected by athletic therapists or team doctors who had knowledge of the study and were educated in administering the SCAT; however, the authors were not present at the time of collection. The authors were therefore unable to account for all extraneous factors. The study is also limited by an athlete’s accuracy in reporting previous concussions on their preseason questionnaire. The generalisability of this study to other sports is limited by the use of only wrestlers, football and ice-hockey players.

Conclusion

The mean total baseline PCSS score for varsity collision sport athletes was 5, although just under half of the athletes scored 0. Women had the highest total baseline PCSS score. Women scored higher than men on their cognitive assessment. Interestingly, previously concussed athletes scored higher than those who were never concussed in all cognitive tests except delayed five-word recall. The “follow-up symptoms only” section of the symptom scale was not completed by 23.5% of athletes on baseline testing, suggesting that the wording of this section is unclear and should be changed in future versions.