The standing knee lift test is not a useful screening tool for time loss from low back pain in youth basketball and floorball players

This is a self-archived version of an original article. This version may differ from the original in pagination and typographic details. Author(s): Title: Year: Version: Copyright: Rights: Rights url: Please cite the original version: CC BY-NC-ND 4.0 https://creativecommons.org/licenses/by-nc-nd/4.0/ The standing knee lift test is not a useful screening tool for time loss from low back pain in youth basketball and floorball players © 2021 Elsevier Accepted version (Final draft) Rossi, Marleena Katariina; Pasanen, Kati; Heinonen, Ari; Äyrämö, Sami; Leppänen, Mari; Myklebust, Grethe; Vasankari, Tommi; Kannus, Pekka; Parkkari, Jari Rossi, M. K., Pasanen, K., Heinonen, A., Äyrämö, S., Leppänen, M., Myklebust, G., Vasankari, T., Kannus, P., & Parkkari, J. (2021). The standing knee lift test is not a useful screening tool for time loss from low back pain in youth basketball and floorball players. Physical Therapy in Sport, 49, 141-148. https://doi.org/10.1016/j.ptsp.2021.01.017 2021 Journal Pre-proof The standing knee lift test is not a useful screening tool for time loss from low back pain in youth basketball and floorball players Rossi Marleena Katariina, Pasanen Kati, Heinonen Ari, Äyrämö Sami, Leppänen Mari, Myklebust Grethe, Vasankari Tommi, Kannus Pekka, Parkkari Jari PII: S1466-853X(21)00043-2 DOI: https://doi.org/10.1016/j.ptsp.2021.01.017 Reference: YPTSP 1342 To appear in: Physical Therapy in Sport Received Date: 30 December 2019 Revised Date: 22 October 2020 Accepted Date: 10 January 2021 Please cite this article as: Katariina, R.M., Kati, P., Ari, H., Sami, Ä., Mari, L., Grethe, M., Tommi, V., Pekka, K., Jari, P., The standing knee lift test is not a useful screening tool for time loss from low back pain in youth basketball and floorball players, Physical Therapy in Sports, https://doi.org/10.1016/ j.ptsp.2021.01.017. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2021 Published by Elsevier Ltd. Title: The standing knee lift test is not a useful screening tool for time loss from low back pain in youth basketball and floorball players Authors (Family name, given name, degrees) Rossi, Marleena Katariina, PT, MSc 1, 2 , Pasanen, Kati, PT, MSc, PhD 1, 3, 4, 5 , Heinonen, Ari, PT, PhD 2 , Äyrämö, Sami, PhD 6, Leppänen, Mari, PhD 1 , Myklebust, Grethe, PT, PhD 7 , Vasankari, Tommi, MD, PhD 1 , Kannus, Pekka, MD, PhD 1, 9 , Parkkari, Jari, MD, PhD 1, 8 Affiliations (Address): 1 Tampere Research Center of Sports Medicine, The UKK Institute for Health Promotion Research, Kaupinpuistonkatu 1, 33500 Tampere, Finland 2 Faculty of Sport and Health Sciences, P.O. Box 35 40014 University of Jyväskylä, Jyväskylä, Finland 3 Sport Injury Prevention Reserch Centre, Faculty of Kinesiology, University of Calgary, 2500 Universtity Drive NW, Calgary, AB, T2N 1N4, Canada 4 Alberta Children's Hospital Research Institute, University of Calgary, 3330 Hospital Drive NW, Calgary, AB, T2n 4N1, Canada 5 McCaig Insitute for Bone and Joint Health, 3280 Hospital Drvie NW, Calgary, AB, T2N 4Z6, Canada 6 Faculty of Information Technology, P.O. Box 35 40014 University of Jyväskylä, Jyväskylä, Finland, 7 Oslo Sports Trauma Research Center, Department of Sports Sciences, Norwegian School of Sport Sciences, Sognsveien 220, 0806 Oslo, Norway 8 Tampere University Hospital, Central Hospital, PO BOX 2000, FI-33521 Tampere, Finland 9 Department of Orthopedics & Traumatology, Central Hospital, PO BOX 2000, FI-33521 Tampere, Finland Journal Pre-proof Twitter: Marleena Rossi (a)MarleenaRossi Mari Leppänen ((a)mari_leppanen Grethe Myklebust (a)MyklebustGrethe Kati Pasanen (a)KatiPasanen Acknowledgements: Sincere thank you to Kari Tokola for indispensable support with the statistical methods. Thanks also for Jussi Hietamo and Teemu Ekola for injury data collection, Irja Lahtinen for collection of training and match data and keeping contact with the participating teams. Corresponding author: Marleena Rossi, MSc, PT Tampere Research Center of Sports Medicine/The UKK Institute for Health Promotion Research Kaupinpuistonkatu 1, 33500 Tampere, Finland PL 30 33501 Tampere, Finland Phone: +358-407283228 email: m[email protected] Journal Pre-proof 1 The standing knee lift test is not a useful screening tool for time loss from low back pain in youth basketball and floorball players Journal Pre-proof 2 ABSTRACT 1 Objectives: The aim of this study was to investigate the association between pelvic 2 kinematics during the standing knee lift (SKL) test and low back pain (LBP) in youth 3 floorball and basketball players. 4 Design: A prospective cohort study. 5 Setting: Finnish elite youth floorball and basketball players. 6 Participants: Finnish elite youth female and male floorball and basketball players 7 (n=258, mean age 15.7±1.8). 8 Main Outcome Measures: LBP resulting in time loss from practice and games was rec-9 orded over a 12-month period and verified by a study physician. Associations between 10 LBP and sagittal plane pelvic tilt and frontal plane pelvic obliquity during the SKL test as 11 measured at baseline were investigated. Individual training and game hours were rec-12 orded, and Cox’s proportional hazard models with mixed effects were used for the 13 analysis. 14 Results: Cox analyses revealed that sagittal plane pelvic tilt and frontal plane pelvic 15 obliquity were not associated with LBP in floorball and basketball players during the 16 follow-up. The hazard ratios for pelvic tilt and pelvic obliquity ranged between 0.93 17 and 1.08 (95% CIs between 0.91 and 1.07 and 0.83 and 1.29), respectively. 18 Conclusions: Pelvic movement during the SKL test is not associated with future LBP in 19 youth floorball and basketball players. 20 Journal Pre-proof 3 Keywords: low back pain, risk factors, prospective study, youth athletes 21 Journal Pre-proof 4 INTRODUCTION 22 Low back pain (LBP) is common in youth and presents with a mean lifetime prevalence 23 of 39% (range 8% to 64%) (Calvo-Muñoz, Gómez-Conesa, & Sánchez-Meca, 2013). LBP 24 in youth results in absence from work or school and interference with normal daily ac-25 tivities and recreational physical activities (Coenen et al., 2017). In Finland, nearly half 26 of youth between 11 and 15 years of age participate in organised sports. Studies ana-27 lysing any association between LBP and physical activity have been inconsistent 28 (Kamper, Yamato, & Williams, 2017), but participation in organised sports might in-29 crease the risk for LBP (Franz, Jespersen, Rexen, Leboeuf-Yde, & Wedderkopp, 2016; 30 Hangai et al., 2010). However, prospective studies investigating the risk factors for LBP 31 in youth sports are limited. To effectively decrease the incidence of LBP in youth ath-32 letes, the risk factors for LBP should be identified. 33 34 LBP prevalence is high in youth floorball (an indoor team sport that resembles floor 35 hockey) and basketball players. In our previous investigation, 44% and 62% of the bas-36 ketball and floorball players, respectively, reported having had LBP within the previous 37 12 months (XXX). Both sports include running, sudden turns and stops, as well as other 38 movements performed on single-leg support. In addition, basketball players perform 39 lots of jumping and landing, and floorball includes a lot of positions and movements 40 with a bended trunk because of the relatively short stick used. 41 42 The standing knee lift (SKL) test has been used to evaluate hip and pelvic stability 43 (Corkery et al., 2014; DiMattia, Livengood, Uhl, Mattacola, & Malone, 2005; 44 Journal Pre-proof 5 Elphinston, 2008; Hardcastle & Nade, 1985). Especially in the LBP population, the SKL 45 test and its modifications are often used in clinics to assess if there are impairments in 46 hip and pelvic movement control (i.e., inability to maintain neutral hip and pelvic 47 alignment), and its use has been suggested as a part of functional screening for ath-48 letes (Elphinston, 2008). Increased pelvic movement, for example, increased pelvic 49 obliquity or tilt during the SKL test, may be because of impaired movement control. In 50 theory, altered movement control in single-leg tasks may result in increased loading 51 and strain in the lower back area (posterior lumbo-pelvic area) in these floorball and 52 basketball players. Indeed, alignment of the lumbo-pelvic area has been shown to be 53 associated with lumbar loading (Bassani, Casaroli, & Galbusera, 2019; Popovich et al., 54 2013). 55 56 Further investigations analysing any association between LBP and movement patterns 57 in sports is needed (O'Sullivan, Smith, Beales, & Straker, 2017). Hence, the overall aim 58 of the current study was to investigate the association between LBP incidence and pel-59 vic kinematics during the SKL test in youth floorball and basketball players. The study 60 objective was to assess whether increased sagittal or frontal plane pelvic movement 61 during the SKL test is a risk factor for future LBP that would result in time loss from 62 sports participation in youth floorball and basketball players. Our hypothesis was that 63 players with increased pelvic movement during the SKL test would have an increased 64 risk for LBP. 65 66 METHODS 67 Journal Pre-proof 12 vidual game and practice hours from the start of the follow-up until the first event of 186 LBP or the end of the follow-up (if no event) were included in the Cox analyses. Data 187 from all eligible players entering the follow-up were included for the time when they 188 participated. 189 190 Univariate analyses were followed by multivariable analyses. Two adjusting variables 191 were used in the multivariable analyses because it has been recommended to have 10 192 events per included variable in the Cox analyses (Peduzzi, Concato, Feinstein, & Hol-193 ford, 1995; Peduzzi, Concato, Kemper, Holford, & Feinstein, 1996). First, we included 194 the following factors into one model: age, sex, body mass index (BMI), nicotine use, leg 195 dominance, family history of LBP and history of LBP. Leg dominance was used as two 196 category variables; the categories ‘left’ and ‘right’ were merged into ‘unilateral leg 197 dominance’ and the category ‘don’t know/both’ into ‘bilateral/unknown leg domi-198 nance’. Then, the factors were dropped one by one from the model based on their sta-199 tistical significance (the factors with the largest p-values were dropped). Finally, a his-200 tory of LBP and leg dominance were entered into the final model because of having 201 the highest statistical significance (smallest p-value). The results are presented as haz-202 ard ratios (HRs), 95% CIs and p-values. The player was considered the unit of analysis, 203 and analyses for right and left legs were performed separately. 204 205 RESULTS 206 Nine basketball and nine floorball teams participated in the current study. Thirty-seven 207 players declined and 403 players agreed to participate in the three-year open cohort 208 Journal Pre-proof 13 study. Seventy-nine percent (n=319) of the players agreed to participate during the 209 third study year (2013–2014). Forty-nine players did not have complete SKL test data, 210 eight players did not participate in the follow-up, and four players reported an ongoing 211 acute unilateral injury at the time of testing and were excluded from the analyses (Fig-212 ure 2). 213 214 215 Figure 2. Study flow of the participating players. *Incomplete SKL test data (no testing 216 data n=29, technical reasons n=16, incorrect performance n=4). 217 218 The baseline player demographics are presented in Table 2. Two-hundred-and-fifty-219 eight players participated in the follow-up and SKL test. The mean, minimum and max-220 Journal Pre-proof 14 imum values for the investigated risk factors are presented in Table 3. There was a 221 small number of players (n=40) with actual pelvic drop movements, and the maximum 222 pelvic drop was 3.5 degrees. 223 224 Table 2. Baseline characteristics (n=258) 225 Variables Basketball Floorball P-value Female (n=61) Male (n=67) Female (n=50) Male (n=80) Age, years (mean, (SD)) 14.4 (1.3) 15.1 (1.8) 17.3 (1.8) 16.9 (1.3) ≤0.001 Height, cm (mean, SD) 168.5 (6.5) 179.2 (10.3) 167.0 (6.0) 177.3 (6.0) 0.633 Weight, kg (mean, SD) 60.9 (8.6) 68.2 (13.8) 62.3 (7.6) 69.2 (8.6) 0.087 BMI (mean, SD) 21.4 (2.7) 21.0 (3.0) 22.3 (2.5) 21.9 (2.2) 0.003 Playing years (mean, SD) 6.6 (2.5) 6.8 (3.0) 7.2 (2.5) 8.8 (3.0) ≤0.001 Training hours * (mean, SD) 170.9 (73.4) 246.8 (134.6) 231.7 (106.4) 257.7 (133.5) 0.010 Game hours † (mean, SD) 7.6 (4.7) 7.5 (3.9) 10.7 (7.4) 10.0 (6.9) 0.001 Body mass index, BMI; SD, standard deviation. p-values shown refer to the t-test/Mann-Whitney test between sports groups, including both sexes. *Team practice hours/season. † Active playing time in games during the season. 226 Table 3. Baseline test results for players with and without LBP during follow-up 227 Outcome No LBP during follow-up # LBP during follow-up (n=32) All players Variables Mean (95% CI) Mean (95% CI) Pvalue Mean (95% CI) Min. value Max value Right leg Peak pelvic anterior tilt, degrees // 9.6 (9.1 to 10.2) 9.3 (7.8 to 10.8) 0.854 9.6 (9.1 to 10.1) 0.7 20.6 Peak pelvic posterior tilt, degrees // -4.3 (-5.1 to -3.5) -4.0 (-6.0 to -2.0) 0.797 -4.2 (- 4.9 to - 3.5) -23.3 9.9 Peak contralateral hike angle, degrees^ 13.8 (13.4 to 14.2) 13.0 (11.9 to 14.1) 0.793 13.7 (13.3 to 14.1) 5.2 22.4 Peak contralateral 1.9 (1.6 to 2.1) 1.5 (0.7 to 2.3) 0.934 1.8 (1.6 -3.5 8.0 Journal Pre-proof 15 drop angle, degrees^ to 2.1) Left leg Peak pelvic anterior tilt, degrees // 9.2 (8.6 to 9.7) 9.4 (7.8 to 10.9) 0.691 9.2 (8.7 to 9.7) -1.7 19.9 Peak pelvic posterior tilt, degrees // -4.7 (-5.5 to -3.9) -4.1 (-6.1 to -2.1) 0.814 -4.6 (- 5.3 to - 3.9) -24.4 9.9 Peak contralateral hike angle, degrees^ 14.2 (13.7 to 14.7) 13.9 (12.8 to 15.0) 0.189 14.1 (13.7 to 14.5) 6.5 27.0 Peak contralateral drop angle, degrees^ 2.2 (1.9 to 2.5) 2.2 (1.5 to 2.9) 0.361 2.2 (2.0 to 2.3) -3.4 8.9 LBP; low back pain, CI; confidence interval # Because of an insufficient number of valid trials (< 2 valid trials), four players were excluded from the right side test and six players from the left side test. Right leg n=222, Left leg n=220. // Positive value in pelvic tilt corresponds to pelvic anterior tilt and a negative value to pelvic posterior tilt. ^Positive value in pelvic obliquity corresponds to contralateral pelvic hike and a negative value to contralateral pelvic drop. 228 Time loss LBP was recorded 39 times during the 12-month follow-up in 35 players. 229 Three of these were direct contact injuries (n=1 sacrum contusion, n=2 low back con-230 tusion) and, hence, were excluded from the analysis. LBP in 78% (n=25) of the players 231 had gradual nontraumatic onset, and 22% (n=7) had acute traumatic onset. Seventy-six 232 percent of the nontraumatic onset and 86% of the acute onset LBP resulted in at least 233 an absence of seven days from normal training (mean (SD) nontraumatic onset LBP: 234 54.5±86.0, acute onset traumatic LBP 72.4±131.8 days). The median absence was 14 235 days, which corresponds to moderate severity (Fuller et al., 2006). The incidence of 236 time loss LBP, including only the first episode of LBP during the follow-up, was 0.5 per 237 1000 player hours. The incidence rate was 12% in the floorball players and 12% in the 238 basketball players. There were no statistically significant differences between the 239 players with and without time loss LBP during the follow-up in the baseline character-240 istics (age, sex, height, weight, BMI, playing years, team training or game hours). 241 Journal Pre-proof 16 242 Risk factor analyses 243 The results from the univariate analyses are shown in Table 4. None of the investigated 244 risk factors were associated with LBP in the univariate Cox analyses. 245 246 Table 4. Unadjusted hazard ratios (HRs) and confidence intervals (CIs) from the Cox 247 mixed-effect analyses. 248 249 In the adjusted Cox regression analysis, no association between sagittal plane pelvic tilt 250 and LBP was found when adjusted for a history of LBP and leg dominance (Figure 3). 251 Furthermore, none of the analyses between pelvic obliquity and LBP revealed signifi-252 Risk factors HR 95 % CI P Left leg Peak pelvic anterior tilt 1.00 (0.92, 1.09) 0.930 Peak pelvic posterior tilt # 0.98 (0.93, 1.05) 0.610 Peak contralateral hike angle 0.98 (0.89, 1.09) 0.710 Peak contralateral drop angle^ 1.01 (0.85, 1.18) 0.950 Right leg Peak pelvic anterior tilt 0.98 (0.90, 1.07) 0.630 Peak pelvic posterior tilt # 0.99 (0.94, 1.06) 0.860 Peak contralateral hike angle 0.94 (0.85, 1.04) 0.250 Peak contralateral drop angle ^ 1.08 (0.90, 1.28) 0.410 HR calculated per one - degree increase . # HR converted so that one-unit increase is interpreted as more pelvic posterior tilt. ^HR converted so that a one-unit increase is interpreted as more pelvic movement towards pelvic drop. Journal Pre-proof 17 cant associations (Figure 4). The peak pelvic drop angle was also analysed as a catego-253 rised risk factor (no pelvic drop = contralateral pelvic drop values at zero or higher, 254 small pelvic drop = contralateral pelvic drop values smaller than zero). The results 255 showed no significant difference in risk between players with or without pelvic drop. 256 257 258 Figure 3. Adjusted hazard ratios (HR) and confidence intervals (CIs) from the Cox 259 mixed-effect analyses with incidence of LBP as the outcome and peak pelvic tilt as a 260 risk factor. Adjusted for history of LBP and leg dominance (unilateral leg domi-261 nance/bilateral leg dominance). 1 HR converted so that a one-unit increase is inter-262 preted as more pelvic posterior tilt. 263 264 0.85 0.9 0.95 1 1.05 1.1 Hazards Ratio Peak pelvic anterior tilt Left leg HR 0.99, 95% CI 0.91 to 1.07 Right leg HR 0.97, 95% CI 0.89 to 1.05 Peak pelvic posterior tilt 1 Left leg HR 0.99, 95% CI 0.93 to 1.06 Right leg HR 1.01, 95% CI 0.95 to 1.07 Journal Pre-proof 18 265 Figure 4. Adjusted hazard ratios (HR) and confidence intervals (CIs) from the Cox 266 mixed-effect analyses with incidence of LBP as the outcome and peak pelvic obliquity 267 as a risk factor. Adjusted for history of LBP and leg dominance (unilateral leg domi-268 nance/bilateral leg dominance). 1 HR converted so that a one-unit increase is inter-269 preted as a smaller minimal value, that is, pelvic movement towards pelvic drop. 270 271 DISCUSSION 272 The current prospective study showed that sagittal plane pelvic tilt during the SKL test 273 is not a risk factor for LBP in youth basketball and floorball players. We observed no 274 association between LBP incidence and sagittal plane pelvic tilt or frontal plane pelvic 275 obliquity during the SKL test, which was in opposition to our hypothesis. 276 277 Our hypothesis was that increased pelvic movement during the SKL test could result in 278 compensatory movement in the low back area and increase the risk for LBP. In theory, 279 increased pelvic movement might lead to increased load and strain in the low back ar-280 0.75 0.80 0.85 0.90 0.95 1.00 1.05 1.10 1.15 1.20 1.25 1.30 Hazards Ratio Pelvic obliquity, Peak contralateral hike angle Left leg HR 0.98, 95% CI 0.88 to 1.09 Right leg HR 0.93, 95% CI 0.84 to 1.04 Pelvic obliquity, peak contralateral drop angle 1 Left leg HR 0.98, 95% CI 0.83 to 1.16 Right leg HR 1.08, 95% CI 0.90 to 1.29 Journal Pre-proof 19 ea. Our hypothesis was based on the widely known kinematic chain theory, where 281 movement in one section affects the other sections of the kinetic chain (Karandikar & 282 Vargas, 2011); this theory is supported by Leppänen et al. (2020), who showed that 283 pelvic hike during the SKL test increases the risk for knee injuries in youth athletes. In 284 addition, it has previously been shown that lower extremity kinematics (Bayne, Elliott, 285 Campbell, & Alderson, 2016) and movement control of the lumbo-pelvic area 286 (Grosdent et al., 2016; Roussel et al., 2009) might be associated with LBP in youth ath-287 letes. For example, Roussel et al. (2009) prospectively investigated the relationship be-288 tween movement control of the lumbo-pelvic area during hip movements and future 289 lower extremity injuries and LBP; they observed an increased risk for lower extremity 290 injuries and LBP in dancers with altered lumbo-pelvic movement control (Roussel et al., 291 2009). Chaudhari, McKenzie, Pan and Oñate (2014) observed increased odds for time 292 loss from a sports injury in baseball pitchers with larger sagittal plane lumbo-pelvic 293 movement during a single-leg raise test in standing. We were unable to find significant 294 risk factors in pelvic kinematics during hip flexion movement in youth basketball and 295 floorball players using the SKL test. Our results are in line with those of Olivier, Stew-296 art, Olorunju and McKinon (2015), who noticed that lumbo-pelvic movement control 297 did not predict injuries in cricket players. 298 299 We hypothesised that increased pelvic obliquity might predispose players to LBP be-300 cause earlier studies suggest that pelvic obliquity can increase facet joint forces and 301 disc pressure (Popovich et al., 2013). However, we did not find an association between 302 pelvic obliquity and LBP. This might be because the data presented only a few and min-303 Journal Pre-proof 20 imal values of pelvic drop. On the other hand, the data show that excessive pelvic drop 304 during the SKL test is not common in youth basketball and floorball players and that 305 the SKL test might not be suitable for detecting players with altered pelvic control dur-306 ing single-leg tasks. 307 308 In the present study, we did not consider that the risk factors for LBP might differ 309 based on many factors, such as tissue injury, onset mechanism, sports-specific re-310 quirements, symptom picture, such as pain provoked by certain movement directions, 311 as well as other characteristics of LBP and the characteristics of the players, such as 312 sex. For example, when investigating LBP—irrespective of the onset or duration of LBP 313 or presence or absence of movement control impairments and provocative movement 314 directions (Astfalck et al., 2010; Dankaerts, O'Sullivan, Burnett, & Straker, 2006)—this 315 so-called ‘wash out’ effect may happen. For example, when investigating nonspecific 316 LBP classified into subgroups based on the presence of movement control impairments 317 and provocative movement directions, differences in movement patterns in people 318 with and without LBP can be seen (Astfalck et al., 2010; Dankaerts et al., 2006; Danka-319 erts et al., 2009). Thus, it might be beneficial to investigate risk factors for LBP in dif-320 ferent kinds of LBP and subgroups. 321 322 Furthermore, LBP complaints are a heterogeneous group, and with most of the com-323 plaints, the exact cause for pain cannot be identified. Also, psychosocial factors affect 324 the pain experience. This makes it more difficult to subgroup LBP based on, for exam-325 ple, injured tissue and, hence, to identify risk factors for LBP complaints. 326 Journal Pre-proof 21 327 For the Cox analysis, we did not enter all adjusting factors available, such as age, sex, 328 BMI and family history of LBP, into the final risk factor analyses, even though prior 329 studies have stated them as plausible predisposing factors for LBP (Ferreira, Beck-330 enkamp, Maher, Hopper, & Ferreira, 2013; Kamper et al., 2017). This was because of 331 applying the rule of 10 incidents per variable in the model (Peduzzi et al., 1995; Peduz-332 zi et al., 1996). We included age, sex, BMI, nicotine use, leg dominance, family history 333 of LBP and history of LBP in the same model, and one by one, we dropped the least 334 significant variables from the model. We noticed that only nicotine use, history of LBP 335 and leg dominance were statistically significant factors. Interestingly, sex, age and BMI 336 were not statistically significant. History of LBP and leg dominance had the lowest p-337 value and were included in the final adjusted analyses. Out of curiosity, we also ran the 338 analyses using nicotine and history of LBP as adjusting factors, but the results re-339 mained the same. 340 341 Strengths and limitations 342 The strengths of this investigation were the 12-month follow-up and prospective regis-343 tration of the individual training hours, game hours and LBP complaints. The sample 344 can also be seen as representative of youth basketball and floorball players of the 345 same level in Finland. 346 347 Despite the strengths, there are also limitations to consider. We did not perform a reli-348 ability analysis of the 3D SKL test. However, one trained physiotherapist performed the 349 Journal Pre-proof 28 Kamper, S. J., Yamato, T. P., & Williams, C. M. (2017). The prevalence, risk factors, 473 prognosis and treatment for back pain in children and adolescents: An overview of 474 systematic reviews. Best Practice & Research: Clinical Rheumatology, 30(6), 1021-475 1036. doi:10.1016/j.berh.2017.04.003 476 Karandikar, N., & Vargas, O. O. (2011). Kinetic chains: A review of the concept and its 477 clinical applications. PM&R, 3(8), 739-745. doi:10.1016/j.pmrj.2011.02.021 478 Kuorinka, I., Jonsson, B., Kilbom, A., Vinterberg, H., Biering-Sørensen, F., Andersson, G., 479 & Jørgensen, K. (1987). Standardised Nordic questionnaires for the analysis of 480 musculoskeletal symptoms. Applied Ergonomics, 18(3), 233-237. 481 doi:10.1016/0003-6870(87)90010-X 482 Leppänen, M., Rossi, M. T., Parkkari, J., Heinonen, A., Äyrämö, S., Krosshaug, T., . . . 483 Pasanen, K. (2020). Altered hip control during a standing knee-lift test is associat-484 ed with increased risk of knee injuries. Scandinavian Journal of Medicine & Science 485 in Sports, 30(5), 922-931. doi:10.1111/sms.13626 486 Olivier, B., Stewart, A. V., Olorunju, S. A. S., & McKinon, W. (2015). Static and dynamic 487 balance ability, lumbo-pelvic movement control and injury incidence in cricket 488 pace bowlers. Journal of Science and Medicine in Sport, 18(1), 19-25. 489 doi:10.1016/j.jsams.2013.10.245 490 Olsen, O., Myklebust, G., Engebretsen, L., & Bahr, R. (2004). Injury mechanisms for an-491 terior cruciate ligament injuries in team handball. The American Journal of Sports 492 Medicine, 32(4), 1002-1012. doi:10.1177/0363546503261724 493 Journal Pre-proof 29 O'Sullivan, P., Smith, A., Beales, D., & Straker, L. (2017). Understanding adolescent low 494 back pain from a multidimensional perspective: Implications for management. The 495 Journal of Orthopaedic and Sports Physical Therapy, 47(10), 741-751. 496 doi:10.2519/jospt.2017.7376 497 XXXX 498 XXX 499 Peduzzi, P., Concato, J., Feinstein, A. R., & Holford, T. R. (1995). Importance of events 500 per independent variable in proportional hazards regression analysis II. Accuracy 501 and precision of regression estimates. Journal of Clinical Epidemiology, 48(12), 502 1503-1510. doi:10.1016/0895-4356(95)00048-8 503 Peduzzi, P., Concato, J., Kemper, E., Holford, T. R., & Feinstein, A. R. (1996). A simula-504 tion study of the number of events per variable in logistic regression analysis. 505 Journal of Clinical Epidemiology, 49(12), 1373-1379. doi:10.1016/S0895-506 4356(96)00236-3 507 Popovich, J. M., Welcher, J. B., Hedman, T. P., Tawackoli, W., Anand, N., Chen, T. C., & 508 Kulig, K. (2013). Lumbar facet joint and intervertebral disc loading during simulat-509 ed pelvic obliquity. The Spine Journal, 13(11), 1581-1589. 510 doi:10.1016/j.spinee.2013.04.011 511 R Core Team. (2016). R: A language and environment for statistical computing [com-512 puter software]. Vienna, Austria: R Foundation for Statistical Computing. 513 Journal Pre-proof 30 Roussel, N. A., Nijs, J., Mottram, S., Van Moorsel, A., Truijen, S., & Stassijns, G. (2009). 514 Altered lumbopelvic movement control but not generalized joint hypermobility is 515 associated with increased injury in dancers. A prospective study. Manual Therapy, 516 14(6), 630-635. doi:10.1016/j.math.2008.12.004 517 Therneau, T. M. (2015). coxme: Mixed effects Cox models. R package 518 [computer software] 519 520 Journal Pre-proof HIGHLIGHTS - The association between LBP and pelvic kinematics in youth floorball and basketball players was investigated in this cohort study. - Three-dimensional movement analysis was used, and pelvic kinematics were calculated from standing knee lift test. - Individual training and game hours and time-loss LBP were recorded during the 12-month follow-up. - Neither pelvic tilt, or obliquity, during standing knee lift test were associated with future LBP in youth floorball and basketball players. Journal Pre-proof Ethical approval Ethics Committee of Pirkanmaa Hospital District (ETL-code R10169). The study was carried out in accordance with the Declaration of Helsinki and the guidelines for good scientific practice. Journal Pre-proof Conflict of Interest None to declare Ethical Statements Ethics Committee of Pirkanmaa Hospital District (ETL-code R10169). The study was carried out in accordance with the Declaration of Helsinki and the guidelines for good scientific practice. Funding The PROFITS -study was financially supported by the Finnish Ministry of Education and Culture, and the Competitive State Research Financing of the Expert Responsibility Area of Tampere University Hospital (Grants 9N053, 9S047, 9T046, 9U044). The corresponding author was supported with research grants from University of Jyväskylä and City of Kangasala (Björqkvist’s fund). Acknowledgements Sincere thank you to Kari Tokola for indispensable support with the statistical methods. Thanks also for Jussi Hietamo and Teemu Ekola for injury data collection, Irja Lahtinen for collection of training and match data and keeping contact with the participating teams. Availability of data and materials The data can not be shared because permission was not asked from the participants or their parents. Authors’ contributions All authors contributed to study concept and design. KP was responsible for conducting the data acquisition. ML was responsible for preparation of the 3D motion capture data. MR was responsible for the main data analysis, interpretation and writing the first draft of the manuscript. KP, AH, SÄ, ML, GM, TV, PK, and JP were significant manuscript revisers. All authors have approved the submitted version of the manuscript. KP is the guarantor. Journal Pre-proof