Incidence, Prevalence, and Severity of and Risk Factors for Hip and Groin Problems in Swedish Male Ice Hockey Players: A 1-Season Prospective Cohort Study

In this prospective cohort study, we asked participating ice hockey players to report hip and groin problems on the Oslo Sports Trauma Research Center Overuse Injury Questionnaire (OSTRC-O)²⁰  biweekly during the regular 2017–2018 season. At the beginning of the season, players completed a baseline questionnaire on hip and groin problems in the previous season and performed hip-muscle strength tests (bilateral adduction and abduction) and the 5-second squeeze test. The Ethics Committee at Lund University approved this study (Dnr 2017/483). We provided the results in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology guidelines (Extension for Sport Injury and Illness Surveillance).²¹ 

We invited all professional (Swedish Hockey League, Allsvenska) and semiprofessional (Division 1) male ice hockey teams in the greater Stockholm area to participate. Of the 12 invited teams, 9 agreed to participate. Before the first measurements, all individual players received written information about the study. On the day of the first measurements, they had the chance to ask questions about the study and supplied written informed consent.

Before the 2017–2018 season, we visited each participating team before a training session to collect baseline data. Players completed an electronic questionnaire that asked for demographic information, playing level (league), playing position, years of ice hockey experience, and the occurrence and duration of hip and groin problems during the previous season. Information was collected separately for hip and groin problems that led to an inability to participate in training and matches (time-loss injuries) and those that affected self-reported hockey performance without preventing participation (non–time-loss injuries).

Subsequently, we conducted a physical test battery consisting of the 5-second squeeze test¹⁸  and bilateral isometric adduction and abduction strength testing. The 5-second squeeze test was performed with the player in a supine position and the assessor's forearm between the player's ankles. The player was asked to produce a 5-second maximal isometric contraction and rate the magnitude of groin pain during the contraction on a numeric pain rating scale from 0 to 10 (0 = no groin pain, 10 = maximal groin pain). The 5-second squeeze test can indicate hip- and groin-related sporting function and impaired adduction and abduction strength in ice hockey players.¹⁹  Bilateral isometric adduction and abduction strength was measured by a single assessor (T.W.) using a handheld dynamometer (model MicroFET 2; Hoggan Health Industries). Both measures, which have been shown to be reliable,¹⁹  were performed in the 5-second squeeze position. For the abduction measure, a fixation belt held the dynamometer in place on the lower leg at the height it was applied during the adduction measure.

Every second week, players retrospectively reported hip and groin problems that occurred during the preceding 14 days on the OSTRC-O. The instrument is a valid method for recording overuse injuries.^20,22  The OSTRC-O assesses the effect of hip and groin pain on a player's participation, training volume, performance, and pain during ice hockey (Figure 1). When players reported a problem, we also asked them about associated time loss (number of missed matches and games) and the onset (gradual or sudden). When players reported the sudden onset of problems, we asked them to specify whether it happened during training or a match. All players were requested to report exposure to match play and training (number of sessions on ice and off ice). We sent the questionnaire to all players via SMS message (Briteback AB, Norrköping). Messages were sent biweekly on Sundays throughout the regular 2017–2018 season (12 messages between September 2017 and February 2018). Nonresponding players received up to 3 reminders after 1 and 2 days.

We considered any report that included at least 1 of the following answers on the OSTRC-O to represent a hip and groin problem (all hip and groin problems):

• Q1.

  Anything but full participation without hip and groin problems;

• 2.

  Any reduction in training volume;

• 3.

  Any level of affected performance; and

• 4.

  Any hip and groin pain experienced during sport participation.

We categorized reported conditions as substantial hip and groin problems if players reported any of the following answers on the OSTRC-O:

1. Moderate or severe reduction in training volume (Q2);

2. Moderate or severe affected performance (Q3);

3. Inability to participate (Q1, Q2, or Q3).

If a player missed at least 1 day of ice hockey participation, we considered it to be a time-loss problem. We counted multiple sudden or time-loss problems reported by the same individual as separate events if they were separated by at least 2 weeks of full ice hockey participation without symptoms or performance deficits.

We calculated the biweekly prevalence of hip and groin problems (all and substantial) with corresponding 95% CIs for proportions (p; symptotic [Wald] method) based on a normal approximation as p ± 1.96 × (p [1 − p] / n). We then determined the average biweekly prevalence with corresponding 95% CIs for the whole season.

We calculated the proportion of players who experienced at least 1 episode of hip and groin problems during the season (all, substantial, and time-loss problems) as well as incidence rates for sudden-onset and time-loss problems in relation to 1000 AEs (number of matches and training sessions) with 95% CIs (normal approximation to the Poisson distribution). Separate logistic regression analyses were performed to explore the odds of experiencing all, substantial, or time-loss problems at some point during the season in relation to preseason factors (time-loss or non–time-loss injury during the previous season, adduction strength, abduction strength, adduction-to-abduction strength ratio, and 5-second squeeze test results). Continuous predictors were categorized into tertiles. Age and years of hockey experience were separately included in the models and evaluated for potential confounding effects according to the change-in-estimate approach.²³  Given that none of the variables resulted in a change in estimate (odds ratio [OR]) of >15%, they were not included in the final analyses.

We invited 12 teams to participate in the study, and 10 teams agreed. At baseline, 212 players underwent testing. One of these 10 teams (18 players) declined prospective follow-up, leaving us with 194 players (17–28 players per team) for prospective injury surveillance. In total, 163 players (84%) provided surveillance data and were included in the final sample. Player characteristics are summarized in Table 1. The baseline questionnaire was completed by all included players, and the average biweekly response rate was 62.7% (minimum = 50.3%, maximum = 85.6%).

Over the course of the season, we recorded a total of 182 problem reports (in 74 unique players), of which 73 (in 31 unique players) were classified as substantial hip and groin problems. In total, 45.4% (95% CI = 37.6%, 53.4%) of all players reported at least 1 episode of hip and groin problems (all problems), and 19.0% (95% CI = 13.3%, 25.9%) reported at least 1 episode of substantial problems. On average, 14.1% (95% CI = 10.8%, 17.5%) of all players experienced hip and groin problems (all problems) during any given 14-day interval, and 5.7% (95% CI = 4.3%, 7.2%) experienced substantial hip and groin problems during the regular season (Figure 2).

Time-loss problems were experienced at least once during the season by 14.7% (n = 24) of players. In total, we recorded 26 separate time-loss problems (in 25 unique players) over the course of the season (incidence rate = 1.3 [95% CI = 0.9, 2.0] per 1000 AEs). Among all reports of hip and groin problems (all problems), 17% led to time loss, and among all reports of substantial hip and groin problems, 34.2% led to time loss.

Among all recorded problems, 31% (n = 56) had sudden onset, and 21% (n = 38) of these were classified as separate problems with sudden onset according to our definition (incidence rate = 1.9 [95% CI = 1.4, 2.7] per 1000 AEs). The incidence rate of problems with sudden onset during a match was 5.5 (95% CI = 3.8, 8.8) per 1000 match-exposures and during training, 1.0 (95% CI = 0.06, 1.7) per 1000 training-exposures. Among all recorded problems, 69% had gradual onset. Among all substantial problems, 62% had gradual onset.

Players who reported a hip and groin problem in the previous season had significantly higher odds of reporting a new one during the season than players without hip and groin problems in the previous season. The odds were 2.6 times higher (95% CI = 1.4, 5.0) for reporting new problems, 4 times higher (95% CI = 1.7, 9.7) for reporting substantial problems, and 3 times higher (95% CI = 1.2, 8.0) for reporting time-loss problems during the season. Non–time-loss conditions in the previous season were associated with the highest odds for new problems. We did not identify any significant associations between preseason hip muscle strength and hip and groin problems during the season (Table 2).

In this study, we used a method design to capture the occurrences and consequences of all problems, not just those leading to time loss, to describe the extent to which male professional and semiprofessional ice hockey players sustained hip and groin conditions. Furthermore, we assessed potential risk factors for both time-loss and non–time-loss problems. Our main findings were that almost half of all players reported a hip and groin problem during the season and that 1 in 6 players had problems at any given time during the season. Fewer than 1 in 5 problems led to time loss, and non–time-loss problems in the previous season were a significant risk factor for new substantial problems.

Approximately 14% of ice hockey players had hip and groin problems and approximately 6% had substantial problems at any given time during the season. In a similar investigation¹⁰  of ice hockey goalkeepers, the corresponding prevalences were almost twice as high, indicating that goalkeepers were the most affected players on a hockey team. According to our results, we can expect approximately 11 players per ice hockey team (of 25 players) to experience a hip and groin problem over the course of a season. Therefore, our prospective study results support recent cross-sectional research⁹  that demonstrated a similarly high seasonal prevalence of hip and groin problems in Swedish professional ice hockey players. When we consider only time-loss problems to be recordable events, as is traditional, the number of recorded hip and groin conditions drops dramatically. Data for more than 7000 National Hockey League players over 6 seasons showed an overall incidence rate of 1 per 1000 AEs.⁵  In our study, more than 20 years later, we found an incidence rate of 1.3 time-loss problems per 1000 AEs. However, time-loss events accounted for only 17% of all problems, leaving the majority of hip and groin problems undetected. This result is consistent with previous studies in ice hockey¹⁰  and other sports.^11,24  Instead of expressing the severity of problems by measuring only the extent of time loss, we also included perceived functional impairment in our definition of a substantial problem.²⁰  At any time during the season, 6% of all players had substantial hip and groin problems, meaning they could not train and play at all or could play only with reduced volume and impaired performance. The fact that only one-third of all substantial problems led to time loss further highlights the importance of looking beyond that definition of injury when describing the severity of hip and groin problems in ice hockey players.

In line with the existing investigations,^14,16  we observed that hip and groin problems during the previous season were a risk factor for new hip and groin problems. However, in contrast to earlier authors, we also included the previous season's non–time-loss problems in our risk factor analysis and determined that they were associated with significantly higher odds for both all and substantial groin problems during the season. According to previous researchers,⁹  functional impairments become more severe the longer these hip and groin symptoms persist. Still, more than 4 of 5 players with groin pain in our sample kept playing, possibly putting themselves at risk for more severe problems in the long term. These results emphasize the need to increase our preventive efforts for players with a history of hip and groin conditions, independent of time loss, which may be more important than preseason strength measures. Preseason hip-muscle strength has been associated with in-season groin injuries in 1 study¹⁵  but not in another study.¹⁴  Adduction force was reduced in ice hockey players with ongoing groin symptoms, and the 5-second squeeze test was a potential tool for identifying players with reduced adductor strength and impaired athletic function.^19,25  We did not find hip muscle strength and pain during the 5-second squeeze test to be significantly associated with increased odds for hip and groin problems. We did note higher odds for problems among players with pain during the 5-second squeeze test, although the increase was not significant. Due to the small number of players with high pain levels during the test and the relatively low number of problems, the CIs for this part of the analysis were wide and the uncertainty in estimates called for a cautious interpretation. Further work is needed to establish whether the 5-second squeeze test is associated with the risk of future problems in ice hockey players.

Previous injury was consistently the most powerful risk factor for groin injury in sports, ¹⁶  and although we would have preferred to have identified more “modifiable” factors, our findings highlight the need to prevent the index injury. In football, groin conditions have been prevented using a single-exercise strengthening program for the adductor muscles.²⁶  The progressive strengthening program in our study was aimed at all players (without identifying individual players deemed at risk for groin problems) and reduced the odds for self-reported groin problems by 41%. Preseason adductor strengthening has also been examined in professional ice hockey players. Instead of aiming the intervention at all players, Tyler et al²⁷  singled out individuals with reduced adductor strength as intervention targets. After the intervention, a significant reduction in adductor strains in comparison with the previous seasons was observed. Whereas preventive efforts should not be focused only on specific groups or individuals but preferably provided to all players, some attention to players considered to be at increased risk may still be warranted. However, our results led to questions about the effectiveness of picking “at-risk players” on the basis of clinical preseason measures such as strength. Instead, it may be more effective to pay attention to players with previous problems and to closely monitor symptoms to identify and manage players who play through groin pain and may incur more severe problems. Players with ongoing problems can be identified and monitored using simple screening tools such as the 5-second squeeze test, which can guide management.¹⁹  Such a secondary preventive approach has been suggested for managing groin pain in football players¹⁷  but needs to be explored prospectively. Primary and secondary prevention efforts could be combined to reduce the burden of hip and groin problems in ice hockey players. The ice hockey season has a long summer break during which adductor strengthening can be implemented for all players to increase load tolerance as is done in football.²⁶  Also, more sport-specific training during the off-season may reduce the risk for new injuries because less sport-specific training during the off-season was a significant risk factor for groin injuries.¹⁴ 

Certain methodologic factors should be considered when interpreting our findings. About 80% of our sample consisted of semiprofessional hockey players (third tier of the Swedish league), limiting the generalizability of the results to professional hockey players, who made up only 20%. We had preseason data for all players, but on average, 62.7% of them responded to the biweekly problem reports. In comparison with other investigators²²  who used the OSTRC-O, our response rate was low, which may have biased the results if players with problems were more prone to report. We collected injury data only during the regular season and hence cannot draw conclusions about the preseason or playoffs. We made that decision in order to address the part of the season in which exposure was similar for all players. The number of substantial hip and groin problems was low, resulting in less precision in the estimated odds ratios, as indicated by the wide CIs. For logistical reasons, we could not further classify groin problems into adductor-, iliopsoas-, inguinal-, or symphysis-related groin pain.²⁸  Many teams did not have medically trained staff, and geographic distances made it impossible to bring players to the clinic for physical examinations. For the same reasons, we did not attempt to either identify when groin pain may have originated from the hip joint or categorize these cases into femoroacetabular impingement syndrome, acetabular dysplasia, or other nonmorphologic conditions, as was recommended in a recent consensus statement.²⁹  Future investigations are needed to further classify hip and groin pain in ice hockey players.

Hip and groin problems are common in ice hockey players and can be associated with substantial impairments in performance. Only 1 in 5 problems led to time loss, and 7 of 10 had a gradual onset. Non–time-loss problems in the previous season were a significant risk factor for new problems, whereas decreased preseason hip-adduction and -abduction strength was not. Our findings highlight the importance of detecting and addressing non–time-loss conditions early and increasing our preventive efforts in all players.