This cross-sectional study was approved by the University Medical Center Utrecht ethics committee (protocol number 16–781/C). The manuscript was written using the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline .
771 adult athletes (≥18 years), with adequate Dutch language skills, who used lower extremity CGs from ‘Herzog Medical’ (Herzog Medical B.V. Woudenberg, the Netherlands), regardless of the level in which they participated in their sports, were eligible for inclusion in our study. We chose to include athletes using CGs from only one manufacturer to ensure that the garments had similar weaves and pressure gradients.
The athletes were recruited in April 2017. First, the manufacturer asked the athletes if they were interested in participating in this study. If so, the researchers continued the study, independently of the manufacturer. Each interested athlete received an information letter with details of the study from the researchers and was invited to participate. Four weeks later, athletes who were willing to participate, received an e-mail containing a hyperlink, which could be used to access the questionnaire in NetQ (NetQuestionnaires, NetQ Healthcare B.V., Amsterdam, The Netherlands). The athletes could only fill in the questionnaire after they provided written informed consent. They had five weeks to fill in the questionnaire. Athletes who had not completed or started the questionnaire were sent a reminder every two weeks.
Characteristics of the athletes
Athletes provided information about their age, sex, body mass, height, and, where appropriate, comorbidities. They also provided information about the sports they practised, training frequency, and participation in competitions in the three months before study inclusion. The three-month period was chosen in order to reduce the risk of recall bias.
For this study athletes were divided into two sports categories: endurance and non-endurance athletes (Table 1). This categorization was adapted from previous studies [16, 17]. Endurance sports are characterized by repeated contractions of large skeletal muscle groups at a submaximal intensity over prolonged periods of time for which the energy is delivered mainly by the aerobic system . The main purpose of endurance sports is to increase endurance performance, i.e. to progressively increase the anaerobic threshold (i.e. the start of anaerobic metabolism towards higher exercise intensity). For the purpose of this study, non-endurance athletes were those athletes who did not report an endurance sport as their primary sport. Questions included in the section on the characteristics of the athletes were adapted from previously published studies [19,20,21].
If athletes reported having a past or current lower extremity sports injury, they were asked about its onset mechanism (i.e. acute or gradual), whether they used their CG for this injury, time since occurrence, and the location, type, and duration of the injury. A sports injury was defined as any self-reported physical complaint deemed by the athletes themselves to be caused by participating in their sport and which rendered the athlete unable to participate (fully) in their sport(s), irrespective of the need for medical attention .
Compression garments (CGs)
The athletes reported which type of CG they used: PRO sports compression socks (pressure; about 30 mmHg around the ankle and about 23 mmHg around the knee), PRO sports compression tubes (pressure; about 30 mmHg of pressure around the ankle and about 23 mmHg of pressure around the knee), active compression garments (pressure; about 28 mmHg around the ankle and about 22 mmHg around the knee), ankle compression socks (pressure; about 22 mmHg on the back of the foot to 25 mmHg in the line from the heel to the instep of the foot), thigh support garments (pressure; about 17 mmHg around the knee to about 10 mmHg around the thigh), PRO knee compressive support garments (pressure; about 18–20 mmHg). The CG pressure gradients were provided by the manufacturer. If athletes reported that they used any other than the aforementioned CGs they were excluded from the analyses.
Furthermore, the athletes were asked, for which reason, and how often they used their CGs. Athletes reported their primary and, if applicable, secondary reason (athletes could select, if applicable, multiple secondary reasons) for using the CGs: 1) primary prevention [i.e. prevention of a sports injury that has not occurred]; 2) secondary prevention [i.e. prevention of recurrence of a sports injury previously experienced by the athlete]; 3) to aid post-exercise recovery [i.e. recovery from a competition or training in such a way that the body is prepared for the next session]; 4) to improve performance [i.e. improvement aimed at achieving a predetermined goal for a sport]; 5) to reduce symptoms of a current sports injury; 6) to look good; 7) no specific reason, and 8) other. If athletes answered ‘other’ they could report other primary or secondary reasons for using CGs. Athletes were also asked to report the frequency of the use of their CGs: 1) during, 2) directly after, or 3) the day after training or competition participation by using the following Likert-scale (percentage of the time they use CGs): ‘never’ (0%), ‘sometimes’ (1–35%), ‘regularly’ (36–75%), ‘often’ (76–95%), or ‘always’ (96–100%).
The athletes that aimed to prevent (re-)injury, reported their perceived effects on primary or secondary injury prevention as ‘strong’, ‘partial’, or ‘no effects’. Athletes who indicated that they used the CG to aid post-exercise recovery, to improve performance, or to reduce symptoms of a current sports injury reported the perceived effects of the indicated reason as ‘positive’, ‘neutral’, or ‘negative’. The authors decided on these descriptors in order to attain rough estimates and directions of the perceived effects.
The questionnaire was piloted before the study started. Based on the results of the pilot (unpublished) questions regarding the reasons for CGs use, sports injuries, and perceived affect were adjusted to the version used in this study. For the sections CG use and sports injuries questions were combined and wording was changed to make questions more specific. For the perceived effect section the answer categories were changed from a percentage score to the aforementioned answer categories used in this study.
All data were analysed using SPSS (version 22, IBM, Armonk, New York, USA.). Athletes were included in the analyses if they completed the personal characteristics section of the questionnaire. Athlete characteristics are reported as means and standard deviations (SD) for continuous data, median and 25–75% interquartile range (IQR) for numerical data that were not normally distributed, and percentage and frequency for categorical data.
In order to see if there were any significant differences in the characteristics of the endurance and non-endurance athletes the Chi-squared test and the Student’s T-test were used for ordinal and continuous variables, respectively. If the continuous variables were not normally distributed, the Mann-Whitney U-test was used.
The prevalence of sports injuries was calculated as the number of reported sports injuries divided by the number of athletes at risk . Additionally, the sports injury incidence rate (the number of injuries per 1000 training hours (95% interval [CI])) during the past three months was calculated as follows: (number of new sports injuries during the past three months/number of athletes at risk)*(1000/hours spent training during the past three months) . For the purpose of this manuscript, only the first reported past and current sports injuries were included in the analysis.
The association between the characteristics of athletes who reported running as their primary sport and who used the PRO compression socks or tubes and the odds of a current lower leg sports injury was investigated using multivariate logistic regression analyses . In order to increase the power of this analyses, only runners wearing compressions socks and tubes were included in these analyses. The following characteristics were included in the analyses: age, sex, body mass index, a lower leg sports injury during the 12 months prior to the study, type of CG used, use of > 1 CG, CG use, training parameters during the last three months, and participation in competitions in the last 12 months. Older age, female sex, a higher body mass index, a lower leg sports injury during the 12 months prior to the study, and a lower average weekly running distance (injured athletes were hypothesized to run a lower weekly distance ) were expected to be associated with a higher odds of a current sports injury [26, 27]. The first multivariate model was adjusted for age and sex only. Based on the first multivariate model, scientific literature, and consensus among the authors, the following variables were included in the full multivariate regression model: age, sex, lower leg sports injury during the 12 months prior to the study, use of CGs, and average running distance per week. In order to test the assumptions of multicollinearity, a tolerance of < 0.1 and variance inflation factor > 10 were used [28, 29]. A priori alpha was set at 0.05.