Risk factors associated with low back pain in rowers: a systematic review and meta-analysis

Abstract

Background

Low back pain (LBP) is a prevalent injury among rowers, impacting their careers and potentially causing disability. However, the risk factors for LBP in rowers are still unclear, with most existing studies focusing on individual factors and lacking comprehensive meta-analyses.

Methods

This systematic review and meta-analysis, following PRISMA guidelines, conducted comprehensive searches in PubMed, Web of Science, Embase, and Medrxiv without publication date restrictions. Pooled odds ratios (OR) and 95% confidence intervals (CI) were calculated using a random-effects model to account for heterogeneity. Sensitivity analyses assessed result robustness by excluding individual studies, and meta-regression explored potential heterogeneity sources. The study protocol was registered in PROSPERO (CRD42024590865).

Results

A total of 2,081articles were retrieved, and 10 studies (N = 2,082) were included in the analysis after screening. These comprised 7 cross-sectional and 3 cohort studies, with an average quality score of 7.4. Meta-analysis results showed that age (OR = 1.05, 95% CI: 0.98–1.12), sex (OR = 1.37, 95% CI: 0.63–2.95), BMI (OR = 1.00, 95% CI: 0.91–1.11), competitive level (OR = 1.45, 95% CI: 0.86–2.45), training volume (OR = 1.26, 95% CI: 0.83–1.91), rowing type (OR = 0.69, 95% CI: 0.21–2.25), and Number of years rowing (OR = 1.02, 95% CI: 0.92–1.13) were not significantly associated with low back pain (LBP) in rowers. In contrast, a history of previous LBP (OR = 2.65, 95% CI: 1.86–3.78) was significantly associated with LBP in this population.

Conclusion

Previous low back pain (LBP) is significantly associated with LBP in rowers. Future work requires more high-quality prospective studies to elucidate risk factors and support evidence-based prevention and rehabilitation strategies for rowers.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13102-025-01153-y.

Introduction

Lower back pain (LBP) is a prevalent musculoskeletal condition among athletes [1], particularly in sports characterized by repetitive trunk movements and substantial loads [2]. This condition not only impairs athletes’ training and competitive performance but can also precipitate the cessation of their athletic careers, imposing long-term health and socioeconomic burdens [3]. Rowing, an intensive full-body sport, exhibits a lifetime prevalence of LBP as high as 51.4% [4], markedly exceeding that observed in other athletic disciplines [5].

Prior research has tentatively explored potential risk factors for LBP in rowers. Some studies suggest that cumulative training duration and volume may contribute to the development of LBP in this population [6], while another investigation highlights subjective leg length asymmetry as a predisposing factor [7]. Additionally, research has indicated that competitive level [8] and psychological factors may precipitate symptom onset [9], particularly among professional rowers. A scoping review identified a history of lower back pain (LBP) and prolonged use of force plates as risk factors for LBP in rowers [10]. Furthermore, variations in age, gender, and body mass index (BMI) have been associated with an increased risk of LBP [11]. However, inconsistencies in study design, sample size, study population (e.g., athletes of varying proficiency), and the definitions and measurement of risk factors have led to discrepant findings across these studies.

Therefore, the objective of this study is to synthesize existing research through a meta-analysis to systematically identify and quantify the primary risk factors associated with LBP in rowers. This analysis aims not only to elucidate the principal risk factors and their effects but also to provide empirical evidence for the formulation of effective prevention and intervention strategies, thereby mitigating the long-term impact of LBP on rowers and safeguarding their athletic careers and enduring health.

Methods

Protocol and registration

The systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and was prospectively registered on PROSPERO (CRD42024590865) [12]. We strictly adhered to the predefined protocol throughout the study.

Search strategy and selection criteria

This review aims to systematically identify studies related to risk factors for low back pain (LBP) in rowers, with a focus on analyzing factors that may increase the risk of LBP in rowers. To this end, a comprehensive search was conducted in four major databases: Web of Science, PubMed, Embase, and MedRxiv. The search strategy utilized keywords such as “rowers,” “low back pain,” and “risk factors,” with detailed search strategies for each database available in the supplementary materials. The search cutoff date was 31 March 2025.

The inclusion criteria were as follows: (1) Study design: cohort studies, case-control studies, or cross-sectional studies; (2) Study population: rowers, regardless of gender, age, or competitive level; (3) Risk factors, protective factors, or prognostic factors for low back pain in rowers; (4) Definition of low back pain: nonspecific low back pain defined as pain or discomfort located between the 12 th rib and the gluteal fold, with or without leg pain; (5) Language of the original studies: only studies published in English were included.

The exclusion criteria were as follows: (1) Low back pain caused by infection, surgery, visceral organ problems, falls, or traffic accidents; (2) Specific low back pain caused by neurological disorders, rheumatoid arthritis, systemic lupus erythematosus, acute traumatic fractures or injuries, spinal cord injuries, tumors, and osteoporosis; (3) Conference papers, reviews, editorials, commentaries, letters to the editor, and animal studies; (4) Duplicate studies;

After collecting references and removing duplicates using Zotero 6, two reviewers (XZ and SZ) independently assessed the titles and abstracts according to the predefined inclusion and exclusion criteria to identify eligible studies. Subsequently, they independently evaluated the full texts of the selected studies, and any disagreements were resolved through discussion with a third researcher (PZ). Finally, all selected studies were rigorously reviewed, and studies with duplicate data were excluded based on detailed information such as study time and location.

Data extraction and quality assessment

After identifying the studies to be included, XZ and SZ independently extracted key information and recorded it in an Excel sheet. All entries were cross-checked for accuracy, and any discrepancies were resolved by another researcher (P.Z). The extracted information included: basic study details (title, authors, year of publication, study design), participant characteristics (sample size, age distribution, sex ratio), and core results (effect sizes of risk factors and 95% confidence intervals), as shown in Table 1.

Table 1.

Characteristics of 10 included studies of riskfactors of LBP

No	Author	Year	Country	Design	N	Median Age(y)	Sex ratio (M/F)	Risk factors	Qualiy score
1.	Y. Tashiro	2015	Japan	Cross-sectional	185	19.8	2.56	①③④⑤⑥⑦⑧	8
2.	Craig Newlands	2015	New zealand	Cohort	76	22.6	1.53	①②④⑤⑨⑩	8
3.	F. Maselli	2015	Italy	Cross-sectional	133	19.0	4.12	①②③④⑥⑦⑪⑫⑬	8
4.	Roald Bahr	2004	Norway	Cross-sectional	199	21.5	1.93	①②⑭⑮	7
5.	Larissa Trease	2020	Australia	Cohort	153	NA	3.67	①	7
6.	Christine M	2016	America	Cross-sectional	249	NA	0.31	①	7
7.	Leo Ng	2013	Australia	Cross-sectional	365	14.8	0.55	①②	7
8.	Ida Stange Foss	2012	Norway	Cohort	173	32.0	1.84	①②⑥⑩	6
9.	Katharina Trompeter	2018	Germany	Cross-sectional	156	22.2	NA	③⑤	8
10.	Tomislav Smoljanovic	2009	China	Cross-sectional	393	18.0	1.14	①	8

Risk factors:(1)Gender(2)Age(3)Rowing type(4)Body Mass Index(5)Competitiveness(6)Training volume(7)Number of years rowing(8)Subjective leg length inequality(9)Rowing discipline(10)History of previous LBP(11)Change of typology of rowing(12)Category(13)Other sports(14)Height(15)Weight

The Newcastle-Ottawa Scale (NOS) was used to assess the quality of the included cohort studies [13]. The scale consists of eight items: (1) representativeness of the exposed cohort; (2) selection of the non-exposed cohort; (3) ascertainment of exposure; (4) absence of outcome at the study start; (5) comparability of exposed and non-exposed cohorts in design and analysis; (6) adequacy of outcome assessment; (7) follow-up duration after outcome occurrence; and (8) adequacy of follow-up for both exposed and non-exposed groups. The maximum score is 9, with ≥ 8 indicating high quality, 5–7 indicating moderate quality, and ≤ 5 indicating low quality.

Cross-sectional studies were assessed using the quality evaluation criteria recommended by the Agency for Healthcare Research and Quality (AHRQ) [14]. The maximum score was 11 points, with each study categorized as good (≥ 8 points), moderate (5–7 points), or poor (≤ 4 points) based on their scores. In our study, two reviewers independently assessed the included studies, and disagreements were resolved through discussion or third-party arbitration. We used Cohen’s Kappa test to determine the level of agreement between the two reviewers. Subsequently, the risk of bias (ROB) in each study was evaluated based on the quality scores of the included literature.

Statistical analysis

Statistical analyses were performed using R software, with the significance level set at P < 0.05. The pooled effect size was represented as an odds ratio (OR) adjusted for confounding factors, along with its 95% confidence interval (95% CI). For studies reporting risk ratios (RR), the results were uniformly converted to OR for pooling. Heterogeneity was assessed using the χ² test and the I² statistic [15]. A fixed-effects model was applied when P ≥ 0.1 and I² ≤ 50%, while a random-effects model was used when P < 0.1 and I² > 50%. Sensitivity analyses were conducted by excluding individual studies one by one to observe changes in effect size and the I² statistic. Additionally, studies with significant differences in outcomes or statistical methods that could not be included in the meta-analysis were summarized narratively.

To address heterogeneity across studies, meta-regression analysis was employed to explore potential sources of heterogeneity [16]. Multivariate meta-regression was conducted using maximum likelihood (ML), and publication bias was assessed by funnel plot visualization and Egger’s test.

GRADE evidence evaluation

The GRADE (Grades of Recommendation, Assessment, Development and Evaluation) system was employed to assess the quality of evidence for each risk factor. The GRADE system categorizes evidence quality into four levels: high, moderate, low, and very low. In the initial assessment, randomized controlled trials (RCTs) are assigned a high quality rating, whereas non-RCTs are assigned a low quality rating. During the grading process, five criteria were used to downgrade the quality of studies: risk of bias, consistency, directness, precision, and publication bias. Additionally, three criteria were used to upgrade the quality of studies: magnitude of effect, dose-response relationship, and impact of residual confounding.

Results

Characteristics of included studies

Search results

A total of 2,081 studies were gathered from four databases using Zotero 6 software. After excluding 1,085 duplicate records, the titles and abstracts of 996 studies were carefully reviewed, and 893 studies that did not meet the inclusion criteria were excluded. Subsequently, the full texts of the remaining 103 articles were retrieved and reviewed, with 93 studies excluded due to incomplete information about study participants, design issues, and other reasons. Ultimately, 10 studies from multiple countries were included for systematic review and meta-analysis. The complete study selection process is shown in Fig. 1.

Fig. 1.

Flow diagram of studies search, selection and inclusion process

Of these 10 studies, 2 were from Australia, 2 from Norway, and the remaining 6 were from China, New Zealand, Italy, Germany, the United States, and Japan, with a total sample size of 2,082 athletes. All studies were published between 2004 and 2020. In terms of study design, the 10 studies included 7 cross-sectional studies and 3 cohort studies, with varying follow-up durations (e.g., median follow-up time, mean follow-up time, or maximum follow-up time).

In the quality assessment, 5 studies scored 8 points and were rated as high quality; 4 studies scored 7 points, and 1 scored 6 points, which placed them in the moderate quality category. No study was rated as low quality, and the average quality score for all included studies was 7.4, as shown in Table 1.

In the cohort studies, the prevalence of rowing-related low back pain ranged from 52.6% to 57%. However, these studies did not clearly distinguish whether the reported prevalence was for 6-month, 12-month, or lifetime prevalence. In contrast, the prevalence of low back pain among rowers in the cross-sectional studies showed a wider range, from 12.04% to 94%. This discrepancy may be related to differences in athlete selection, the definition of low back pain, and the competition level of the athletes.

For example, some cross-sectional studies were limited to collegiate-level athletes, while others may have included professional rowers. Due to the higher training intensity of professional athletes, their risk of developing low back pain is significantly increased. Furthermore, the included cohort studies focused on international elite rowers, making their findings somewhat more representative.

Meta-analysis results of risk factors

This study extracted 15 potential influencing factors from the included literature and conducted a meta-analysis on 8 risk factors mentioned in three or more studies, as shown in Fig. 2. Descriptive analyses were performed for other factors that could not be combined due to insufficient data. A total of 9 cross-sectional and cohort studies investigated the impact of gender on low back pain (LBP) in rowers. The results indicated significant statistical heterogeneity among the studies (I² = 93%, P < 0.01), thus a random-effects model was employed for the meta-analysis. The combined analysis showed that while there was some association between gender and low back pain in rowers, it did not demonstrate a significant effect (Odds Ratio [OR] = 1.37; 95% Confidence Interval [CI]: 0.63–2.95).

Fig. 2.

Summary of the forest plot analysis of risk factors for low back pain in rowing athletes. Note: a Gender; b Body Mass Index; c Competitiveness; d Rowing type; e Age; f Training volume, g Number of years rowing, h a history of previous low back pain

Four studies reported on the impact of age on low back pain (LBP) in rowers. The results revealed significant statistical heterogeneity among the studies (I² = 76%, P < 0.01), leading to the use of a random-effects model for the meta-analysis. The combined analysis indicated that while there was some correlation between age and low back pain in rowers, it did not establish a significant effect (Odds Ratio [OR] = 1.05; 95% Confidence Interval [CI]: 0.98–1.12). Additionally, three studies examined the effect of BMI on low back pain in rowers. The results showed no significant statistical heterogeneity among the studies (I² = 0%, P = 0.92), thus a fixed-effects model was used for the meta-analysis. The combined results indicated that the association between BMI and low back pain was also not significant (Odds Ratio [OR] = 1.00; 95% Confidence Interval [CI]: 0.91–1.11).

Excluding the previously mentioned demographic and physical measurement factors, three studies investigated the impact of competitive level on low back pain (LBP) in rowers. The results revealed significant statistical heterogeneity among the studies (I² = 78%, P < 0.01), necessitating the use of a random-effects model for the meta-analysis. The combined results indicated that while there was some correlation between competitive level and low back pain, its significance could not be established (Odds Ratio [OR] = 1.45; 95% Confidence Interval [CI]: 0.86–2.45). Furthermore, three studies analyzed the impact of training volume on low back pain, and the results also demonstrated significant statistical heterogeneity (I² = 74%, P = 0.02), leading to the use of a random-effects model. The combined analysis suggested a correlation between increased training volume and elevated risk of low back pain (Odds Ratio [OR] = 1.26; 95% Confidence Interval [CI]: 0.83–1.91), although it was not significant.

Concurrent with this, three studies examined the effect of rowing type on low back pain (LBP) in rowers. The results indicated significant statistical heterogeneity among the studies (I² = 83%, P < 0.01), leading to the application of a random-effects model for the meta-analysis. The combined results suggested that rowing type may act as a protective factor against low back pain, but did not reach statistical significance, which may be attributed to insufficient sample size or differences in study design (Odds Ratio [OR] = 0.69; 95% Confidence Interval [CI]: 0.21–2.25).

Two studies investigated the effect of Number of years rowing on low back pain in rowers. The results showed no significant statistical heterogeneity between studies (I² = 36.2%, P = 0.2107), and the association between years of rowing experience and low back pain was not statistically significant (Odds Ratio [OR] = 1.02; 95% Confidence Interval [CI]: 0.92–1.13).

Two studies examined the effect of a history of previous low back pain on low back pain in rowers. The results indicated low statistical heterogeneity between studies (I² = 34.4%, P = 0.2169), and a history of previous low back pain was significantly and positively associated with current low back pain (Odds Ratio [OR] = 2.65; 95% Confidence Interval [CI]: 1.86–3.78).

Additionally, factors such as subjective leg length discrepancy, rowing discipline, changes in rowing posture, classification, other athletic experiences, height and weight were also associated with the occurrence of low back pain in rowers. However, the current data are insufficient to support the combination of effect sizes for these factors, necessitating further validation of their potential as risk factors for low back pain.

Sensitivity analysis

In the analysis of combined risk factors, gender, age, competitive level, and rowing type, Training volume, all showed high heterogeneity. Therefore, sensitivity analysis using the leave-one-out method was conducted for these five factors to explore the sources of heterogeneity, as shown in Fig. 3. Results showed that in the gender factor analysis, only the study by Leo Ng explained 11% of the heterogeneity. For the age factor, the leave-one-out analysis did not sufficiently resolve the heterogeneity. After excluding the study by Roald Bahr et al., I² decreased from 76% to 65%, indicating that this study partially explained the heterogeneity. However, after excluding another study, I² increased from 76% to 84%, suggesting that the study might have masked other unidentified sources of heterogeneity. The results indicate that the partial heterogeneity in gender and age as risk factors for low back pain in rowing athletes can be explained by related studies but has not been fully resolved.

Fig. 3.

Forest plot summary of sensitivity analysis. Note: a Competitiveness; b Rowing type; c Training volume; d Gender; e Age

The heterogeneity in competitive level, rowing type, and training volume can all be explained through leave-one-out analyses. In the analysis of competitive level, the study by Y. Tashiro et al. was identified as the primary source of heterogeneity. After excluding this study, I² decreased to 12%, indicating a significant reduction in heterogeneity. In the analysis of rowing type, the study by Katharina Trompeter et al. was also identified as a major source of heterogeneity. After excluding this study, I² dropped from 83% to 0%, indicating the complete elimination of heterogeneity. For the factor of training volume, the study by Ida Stange Foss was identified as the primary source of heterogeneity. After excluding this study, I² decreased from 74% to 5%, with most of the heterogeneity eliminated.

Meta-regression

To further investigate the heterogeneity of gender as a risk factor for low back pain in rowers, this study employed meta-regression analysis, selecting average age, study design, and athlete type as moderating variables, and applied maximum likelihood methods for multivariate meta-regression. The results indicated that study design did not have a significant impact on the outcomes, while age and athlete type showed significant associations with gender factors. Specifically, age exhibited a significant negative correlation with the relative risk associated with gender factors (p= 0.0465), with a coefficient of − 0.1360. Athlete type also demonstrated a significant negative correlation (p= 0.0166), with a coefficient of − 1.2562. The adjusted R² value was 78.10%, indicating that the included variables largely accounted for the heterogeneity of gender as a risk factor for low back pain in rowers (see Table 2).

Table 2.

Results of meta-regression for the gender

Covariate	Meta-regression coefficient	95% Confidence interval	t	p	Adj R-squared
Design	0.1062	− 1.1056 to 1.3180	0.1718	0.8636	78.10%
Age	− 0.1360	− 0.2699 to − 0.0021	− 1.9910	0.0465
Athlete type	− 1.2562	− 2.2837 to − 0.2286	− 2.3961	0.0166

P P value of meta-regression

Publication bias

To assess potential publication bias, we conducted funnel plot analysis and Egger’s regression test using the gender factor (K= 9). The results indicated that the funnel plot did not show significant asymmetry, and the Egger’s regression test yielded a p-value of 0.0974, suggesting that no significant publication bias was detected, as shown in Fig. 4.

Fig. 4.

Publication-bias funnel plot

GRADE evidence evaluation

In this study, we used the GRADE system to assess the quality of evidence for risk factors related to low back pain in rowers. The results showed that the evidence quality for age, gender, and history of low back pain was low, mainly due to study limitations and insufficient precision of the results. The evidence quality for body mass index, competitive level, rowing type, training volume, and years of rowing was very low. These factors not only had serious study limitations but also showed inconsistency and imprecision in the results. Overall, the current evidence quality was low (See Supplementary Material 3, Table S1.)

Discussion

Low back pain (LBP) is the most common injury in rowing, annually affecting 25% to 81% of athletes and accounting for 15% to 25% of all injuries [17]. This study systematically assessed the risk factors for LBP in rowers through a meta-analysis. The primary findings are as follows: (1) A positive correlation was established between a history of previous low back pain and the occurrence of LBP in rowers; (2) No significant associations were detected between LBP and factors such as gender, age, body mass index (BMI), competitive level, training volume, years of training, and type of rowing. The finding of this study will offer clinicians a certain degree of scientific basis, enabling them to formulate more targeted personalized intervention plans, thereby effectively enhancing the overall health and athletic performance of rowers.

A history of previous low back pain is positively correlated with low back pain in rowers, a finding that is consistent with prior studies. Evidence suggests that a history of previous low back pain is a significant risk factor for future low back pain in rowers [18]. Once an athlete has experienced a lumbar injury, the risk of sustaining a similar injury in the future is significantly increased. This phenomenon is not limited to rowers; other studies have also noted that a history of lumbar injuries is a risk factor for recurrent back injuries in college athletes [19]. Moreover, research has found that athletes with a history of back pain are more likely to experience recurrent back pain, although this does not necessarily lead to the termination of their athletic careers [8]. This may be due to the accelerated degeneration of intervertebral discs and decreased spinal stability caused by mechanical injuries, and previous injuries may lead patients to alter their postures or movement patterns, thereby increasing the burden on certain body parts [20].

However, in this study, we found no significant association between age, gender, or body mass index (BMI) and low back pain in rowers. Previous research has typically regarded these factors as key risk elements for low back pain. For instance, some studies have indicated that male rowers are more likely to report low back pain, whereas female athletes tend to experience more intense pain [21]. Additionally, adult rowers are more prone to low back pain [22]. Among adolescent rowers, the incidence of low back pain increases with age, peaking between 15 and 18 years [21]. Moreover, no significant link was found between BMI and low back pain in rowers. Although higher weight often increases spinal load and low back pain risk in the general population [23, 24], this may not apply to rowing, a sport emphasizing energy conversion and coordination rather than weight classes. Overweight athletes, who struggle with high-intensity training, may exit the sport early, underrepresenting this group in studies.

Apart from demographic factors, this study found no statistically significant associations between low back pain in rowers and factors such as training volume, years of training, rowing type, or competition level. However, previous research has identified training volume [18] and competition level as important contributors to low back pain [25, 26]. Studies have indicated that the risk of low back pain increases with prolonged training duration [22], particularly with extended use of rowing ergometers [27]. Another study revealed that elite rowers are more likely to experience low back pain compared to their lower-level counterparts [28, 29]. This may be attributed to the close relationship between high training volume and low back pain, although this association has yet to be clearly established due to considerable variability in how training volume is measured across studies (e.g., weekly training hours, annual total training time, training intensity). Regarding competition level, the inconsistent results may stem from substantial variability in the athletic levels of the included participants, with differences in training intensity and methods across samples potentially being a major contributing factor. Future research should further examine disparities between athletes at different competitive levels, especially in terms of training load and technical execution, to better understand the impact of competitive level on low back pain. Moreover, studies suggest that sweep rowers, due to greater lumbar lateral flexion and rotation, have a higher incidence of low back pain, especially those engaged in both sweep and scull rowing [30], who are at even greater risk; in contrast, scull rowers experience relatively lower lumbar spine loading and thus a reduced risk of low back pain [31]. Although rowing type may have a protective effect against low back pain, the statistical results were not significant and showed high heterogeneity. Further investigation is needed to elucidate the specific impact of rowing mechanics on lumbar spine loading. A cross-sectional study also found an association between perceived leg length discrepancy and low back pain in rowers [7]. However, due to the small sample size and limitations in study design, no causal relationship can be established. It remains unclear whether the perceived leg length asymmetry arises from abnormal lumbopelvic movement patterns during rowing or serves as a potential contributing factor to persistent symptoms [32].

In addition, this study employed meta-regression analysis to explore the role of sex as a source of heterogeneity in the risk of low back pain among rowers. The results indicated that study design had no significant impact on the findings, whereas age and athlete type were significantly negatively correlated with sex. As age increases, the sex-related risk of low back pain decreases, which may be associated with hormonal changes and the accumulation of training experience [33]. Younger female athletes are more prone to low back pain due to differences in pelvic structure and muscle strength. Furthermore, athlete type (elite vs. amateur) also influences the relationship between sex and low back pain [34]. Although elite athletes train at higher intensities, they benefit from more professional health management, leading to a smaller impact of sex differences on low back pain. In contrast, among amateur athletes, females experience a higher incidence of low back pain, likely due to a lack of professional guidance, suboptimal technique, and inadequate load management [35].

Strengths and limitations

The methodological strengths of this study include adherence to the PRISMA statement and the use of the Cochrane GRADE guidelines for rigorous assessment of the quality of evidence. The main limitations of this study are that only observational studies were included, there was considerable heterogeneity among studies, and the sample sizes were relatively small. In addition, this study did not include commonly used, peer-reviewed databases such as CINAHL and SportDiscus, which are closely related to the fields of health and sports sciences. This may have affected the comprehensiveness of our findings to some extent. Another important limitation is the relatively small number of studies included in the meta-analyses. Typically, meta-regression and funnel plot analyses are more statistically powerful and reliable when at least 10 studies are available. However, due to the limited number of eligible studies, these analyses were conducted in an exploratory manner. As a result, the findings of this study may be more susceptible to potential bias.

Conclusion

This study found that a history of previous low back pain (LBP) is significantly positively correlated with low back pain in rowers. In contrast to other sports, the long-observed BMI factor does not show a significant association with low back pain caused by rowing. However, there is currently a relative scarcity of original research focusing on low back pain in rowing athletes, and the effects of some risk factors have not yet reached a consensus. Future high-quality prospective studies are needed to clarify the risk factors for low back pain in rowing athletes, in order to promote evidence-based prevention and rehabilitation of low back pain in rowing athletes.

Abbreviations

LBP

Low back pain

NOS

Newcastle-Ottawa Scale

AHRQ

Agency for Healthcare Research and Quality

ROB

Risk of bias

OR

Odds Ratio

RR

Risk Ratio

BMI

Body Mass Index

95%CI

95% confidence interval

Authors’ contributions

Topic selection, (X.Z.,S.Z. and P.Z); Registration of the project, (P.Z.); Pilot study, (X.Z. and P.Z.); Data collection (X.Z.,P.Z.and S.Z.); Quality assessment (X.Z.,P.Z.and S.Z.); Data analysis (X.Z. and P.Z.); Data and results verification (X.Z.,P.Z.,X.N.L.); Writing the original draft (X.Z.,P.Z.); Manuscript revision (X.Z.,P.Z.and X.N.L.); Funding (X.Z.)

Funding

None

Data availability

All data and material reported in this review and meta-analysis were from peer-reviewed publications. Data is provided within the manuscript or supplementary information files.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.