Exercise interventions for rotator cuff-related shoulder pain in middle-aged and older adults: a systematic review and meta-analysis Authors: Paulino Vico-Rodríguez, Agustín Aibar-Almazán, Fidel Hita-Contreras, María Auxiliadora Márquez-Apolinario, Julia Gámez-Iruela, Marta Cano-Orihuela, Yolanda Castellote-Caballero Published: 2026 ABSTRACT AbstractBackground/objectivesRotator cuff-related shoulder pain is highly prevalent in middle-aged and older adults and is associated with pain, functional limitations, and reduced quality of life. This systematic review and meta-analysis aimed to evaluate the effectiveness of exercise-based interventions on pain, function, range of motion (ROM), and strength in middle-aged and older adults with rotator cuff-related shoulder pain.MethodsA systematic review and meta-analysis were conducted according to PRISMA 2020 guidelines and registered in PROSPERO (CRD420251236625). Searches were performed in PubMed, Scopus, Web of Science, PEDro, and Cochrane Library databases for studies published within the last 10 years. Randomized controlled trials including adults with rotator cuff-related shoulder pain were included. A total of 1,152 participants across 21 studies were analyzed. Exercise-based interventions primarily targeted the rotator cuff and scapular stabilizer muscles through strengthening, stretching, stabilization, and neuromuscular exercise programs, and were compared with usual care, other exercise approaches, or adjunct therapies. Functional outcomes were mainly assessed using validated instruments such as SPADI, DASH/QuickDASH, and the Constant-Murley Score. Effect sizes were calculated using standardized mean differences, and heterogeneity was assessed with the I² statistic.ResultsTwenty-one randomized controlled trials were included. Exercise-based interventions produced significant improvements in pain (g = -1.454, p < 0.001), function (g = 0.457, p < 0.001), and ROM (g = -2.477, p < 0.001). Improvements in strength were smaller and did not reach statistical significance (g = 0.229, p = 0.085). Although strengthening exercises showed positive trends across studies, the pooled effect on strength was not statistically significant. High heterogeneity was observed (I² = 95.96%), likely due to differences in intervention protocols, participant characteristics, and outcome assessment methods.ConclusionsExercise-based interventions are effective in reducing pain and improving function and ROM in middle-aged and older adults with rotator cuff-related shoulder pain. However, effects on strength appear more limited. Further high-quality studies using standardized exercise protocols and longer follow-up periods are needed to optimize rehabilitation strategies.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12891-026-10035-4.Keywords: Rehabilitation, Shoulder injuries, Rotator cuff injuries, Exercise therapy, Shoulder pain, Meta-analysis Introduction Shoulder pain is one of the most prevalent musculoskeletal disorders in the adult population and is a leading cause of visits to primary care, physiotherapy, and rehabilitation services [1]. Epidemiological estimates indicate that up to a quarter of the population may experience shoulder pain at some point in their lives, and approximately one-third of cases progress to persistent or recurrent forms that limit functionality, work performance, and quality of life [2]. Its prevalence increases with age, making this problem a significant source of disability in older adults, who are more susceptible to degenerative changes and have a reduced capacity for physical recovery [3]. Among the most frequent causes, shoulder pain related to the rotator cuff, which includes tendinopathies, subacromial impingement syndrome, partial or degenerative tears, and mechanical alterations of the joint complex, accounts for up to 70% of clinical diagnoses [4]. These conditions are often associated with a combination of degenerative processes, mechanical overload, strength deficits, impaired neuromuscular coordination, and alterations in scapulohumeral kinematics [5]. The interaction between these factors can lead to persistent pain, loss of mobility, and decreased functional capacity, significantly impacting activities of daily living, especially those involving arm elevation or overhead movements [6]. Therapeutic exercise is considered the first-line treatment for the conservative management of rotator cuff-related shoulder pain, as recommended by numerous international clinical guidelines [7]. Various exercise modalities, such as rotator cuff and scapular muscle strengthening, eccentric and concentric training, motor control exercises, scapular stabilization, specific stretches, and structured home programs, have shown potential for improving shoulder pain, function, and mobility [8, 9]. In addition, some studies have evaluated the use of complementary techniques such as manual therapy, joint mobilization, electromyography biofeedback, and body awareness interventions as a complement to exercise [10]. Despite the wide availability of exercise-based therapeutic options, the literature presents considerable heterogeneity in the protocols used, the intensity and progression of loads, session supervision, intervention duration, and participant characteristics [11]. This variability makes it difficult to identify which exercise programs are most effective in addressing the various manifestations of rotator cuff-related shoulder pain. Furthermore, although numerous clinical trials have been published, many of them primarily include middle-aged adults, while the older population, in whom degenerative changes and the presence of comorbidities can influence treatment response, remains underrepresented in the scientific literature [12, 13]. Recent systematic reviews have evaluated exercise-based interventions for rotator cuff-related shoulder pain, particularly regarding the effectiveness of specific exercise modalities and exercise dosage parameters. For example, Lafrance et al. [12] analyzed the efficacy of exercise therapy according to the FITT principle, while more recent reviews have explored the comparative effects of specific versus general exercise programs in individuals with rotator cuff-related shoulder pain. However, previous reviews have generally included broad adult populations, focused on selected intervention types, or combined exercise with multimodal conservative approaches. In contrast, the present systematic review specifically focuses on middle-aged and older adults, a population with higher prevalence of degenerative rotator cuff disorders and distinct rehabilitation needs. Furthermore, this review provides a comprehensive synthesis of multiple exercise-based interventions and their effects on pain, function, range of motion, and strength.This lack of clarity regarding the effects of different exercise approaches limits healthcare professionals’ ability to select evidence-based interventions tailored to the specific needs of each patient [14]. Furthermore, the diversity of outcome measures used across studies, including pain, function, mobility, strength, scapular kinematics, and quality of life, makes it difficult to synthesize the overall effects of therapeutic exercise on this condition and to directly compare the effectiveness of different exercise protocols [15]. In this context, a rigorous systematic review is needed to synthesize the available evidence on exercise-based interventions for managing rotator cuff-related shoulder pain in middle-aged and older adults. The aim of this review is to analyze the different exercise programs used in clinical trials and evaluate their effects on pain, function, range of motion, strength, and other clinically relevant outcomes in this age-specific population. Given the substantial heterogeneity in exercise protocols and outcome reporting across studies, this review primarily aims to provide a comprehensive synthesis of the current evidence and identify areas requiring further standardization and research. This synthesis may help support clinical decision-making and guide the development of future rehabilitation strategies for middle-aged and older adults with rotator cuff-related shoulder pain. Materials and methods The review process was conducted following the guidelines of the PRISMA 2020 statement [16] and in accordance with the protocol previously registered in PROSPERO (CRD420251236625). Furthermore, the methodological guidelines outlined in the Cochrane Handbook for Systematic Reviews of Interventions [17] were adopted. The electronic database search was completed on December 21, 2025. Eligibility criteria For this systematic review and meta-analysis, predefined eligibility criteria were established. Studies were considered eligible if they included middle-aged and older adults diagnosed with rotator cuff-related shoulder pain and evaluated exercise-based interventions as the main component of treatment. Exercise interventions included strengthening, stretching, stabilization, neuromuscular, motor control, aerobic, or therapeutic exercise programs, either applied alone or combined with adjunct interventions, provided that exercise remained the primary therapeutic component under investigation. Only randomized controlled trials published within the last 10 years were included. Eligible studies were required to report quantitative outcomes related to pain, function, strength, range of motion, quality of life, or other clinically relevant outcomes. Bibliographic search strategy Electronic searches were conducted in the following databases: PubMed, EMBASE, Web of Science, Scopus, CINAHL and Cochrane Central Register of Controlled Trials (Supplementary Material 1). The search strategy combined terms related to the population, the intervention, and the study design. Boolean operators (AND/OR) and truncation were used when necessary to optimize record retrieval: (“Rotator Cuff” OR “Rotator Cuff Injuries” OR “Subacromial Impingement Syndrome” OR “Rotator cuff-related shoulder pain” OR “Shoulder Pain”) AND (“Exercise Therapy” OR strengthening OR “resistance training” OR “neuromuscular training” OR proprioception* OR “physical therapy”) AND (“Aged” OR older adults OR “middle-aged adults” OR “older people” OR seniors) AND (randomized controlled trial OR clinical trial OR RCT). Data extraction process Data extraction was performed using a standardized procedure to ensure the consistency, reliability, and transparency of the process. Two reviewers independently extracted data in duplicate, using a template previously designed and piloted specifically for this systematic review. Any discrepancies between reviewers were resolved through discussion and, when necessary, by consulting a third researcher. All retrieved records were imported into Mendeley Reference Manager (Elsevier, Amsterdam, The Netherlands) for reference management and duplicate removal prior to the screening process. The data extraction template included information corresponding to four main categories: study characteristics, participant characteristics, details of the interventions and comparators, and reported outcomes. Regarding study characteristics, data were collected on the author, year of publication, country, study design, sample size, and duration of follow-up. For the participants, information was extracted on age, sex, diagnostic criteria used, duration of pain, and other relevant clinical variables. Information on the interventions included the type of exercise program, frequency, intensity, duration, specific components (strengthening, motor control, functional exercises, neuromuscular training, proprioception, etc.), as well as comparator characteristics. For the outcomes, measures of pain, function, quality of life, range of motion, muscle strength, and any other relevant clinical outcomes were collected. When studies provided multiple instruments for the same outcome, the primary measure was selected based on the authors’ recommendations or the one most commonly used in the literature. For quantitative synthesis, numerical data including sample size, means, standard deviations, confidence intervals, and post-intervention or change scores were extracted whenever available to calculate effect sizes for the meta-analysis. Outcome measures The primary outcomes included pain intensity and shoulder-related function/disability. Secondary outcomes included range of motion, muscle strength, psychosocial variables, and adverse events when reported. Outcome measures were extracted using validated assessment tools reported by the included studies, such as the Visual Analogue Scale (VAS), Numeric Pain Rating Scale (NPRS), Shoulder Pain and Disability Index (SPADI), Disabilities of the Arm, Shoulder and Hand questionnaire (DASH/QuickDASH), and Constant-Murley Score. Outcomes were categorized according to follow-up duration. Short-term outcomes were defined as assessments performed within 12 weeks after the intervention, whereas long-term outcomes were defined as assessments performed after 12 weeks. When available, data were analyzed separately according to short-term and long-term follow-up periods to explore potential differences in treatment effects over time. Assessment of methodological quality The methodological quality of the included studies was assessed using the PEDro scale [18], a tool widely used in physiotherapy and movement science to evaluate the internal rigor and quality of clinical trials. The PEDro scale consists of 11 items that examine key aspects of experimental design, such as randomization, allocation concealment, blinding of participants, therapists, and evaluators, intention-to-treat analysis, and the presentation of appropriate statistical data. Of these 11 items, 10 contribute to the total score (0–10 points), as the first item refers to external validity criteria and is not scored. Two reviewers independently assessed the methodological quality of each study using the PEDro scale. Discrepancies between the two reviewers were discussed until a consensus was reached, and if no resolution was reached, a third reviewer was consulted. For the interpretation of the scores, the criteria commonly accepted in the literature were adopted: (i) 9–10 points: excellent methodological quality; (ii) 6–8 points: good quality; (iii) 4–5 points: acceptable quality; iv) ≤ 3 points: low quality. Risk of bias assessment The risk of bias of the included randomized controlled trials was assessed using the revised Cochrane Risk of Bias tool for randomized trials (RoB 2), in accordance with the recommendations of the Cochrane Handbook for Systematic Reviews of Interventions. The RoB 2 tool evaluates five domains of bias: bias arising from the randomization process, bias due to deviations from intended interventions, bias due to missing outcome data, bias in measurement of the outcome, and bias in selection of the reported result. Two reviewers independently assessed the risk of bias for each included study. Each domain was classified as “low risk of bias,” “some concerns,” or “high risk of bias.” Disagreements between reviewers were resolved through discussion and consensus, and when necessary, a third reviewer was consulted. An overall risk-of-bias judgment was subsequently assigned to each study according to the RoB 2 guidance. Certainty of evidence assessment The certainty of evidence for the main outcomes was assessed using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach. The certainty of evidence was evaluated across the following domains: risk of bias, inconsistency, indirectness, imprecision, and publication bias. Randomized controlled trials were initially considered as high-certainty evidence and were downgraded when important limitations were identified in one or more domains. The certainty of evidence for each outcome was categorized as high, moderate, low, or very low. Data synthesis and statistical analysis A quantitative meta-analysis was performed using Comprehensive Meta-Analysis software (CMA, Version 4; Biostat, Englewood, NJ, USA). Continuous outcomes were analyzed using standardized mean differences (Hedges’ g) with 95% confidence intervals (CIs), due to the variability in outcome measures across studies. Hedges’ g was selected to account for differences in measurement scales and to reduce small-sample bias. Effect sizes were interpreted according to Cohen’s criteria as small (0.2), moderate (0.5), and large (0.8). When studies used the same measurement scale, mean differences (MDs) were calculated. When different instruments were used to assess the same construct (e.g., function or pain), outcome measures were standardized using standardized mean differences (Hedges’ g). For scales with opposite scoring directions, scores were adjusted when necessary to ensure that all effect sizes were interpreted consistently, with negative or positive values reflecting the same clinical direction across studies. Statistical heterogeneity was assessed using Cochran’s Q statistic, the I² statistic, and Tau-squared (τ²). I² values of approximately 25%, 50%, and 75% were interpreted as low, moderate, and high heterogeneity, respectively. Given the expected clinical and methodological heterogeneity among the included studies, random-effects models were preferentially applied. Where sufficient data were available, subgroup analyses were conducted according to intervention characteristics and outcome categories. Publication bias was explored through visual inspection of funnel plots and statistically assessed using Begg and Mazumdar’s rank correlation test and Egger’s regression intercept test. Duval and Tweedie’s trim-and-fill method and Orwin’s fail-safe N were additionally used when appropriate. Statistical significance was established at p < 0.05. Results Study selection process The study selection process was conducted following the PRISMA 2020 flowchart guidelines. After running the search strategy across all electronic databases, 699 records were initially identified. Following the removal of duplicates using the reference manager, 423 studies remained for screening by title and abstract. In this first phase, 366 records were excluded for not meeting the eligibility criteria (e.g., irrelevant population, lack of exercise-based intervention, non-experimental design, or outcomes unrelated to shoulder pain). Subsequently, 57 potentially relevant articles were assessed in full text. Of these, 37 studies were excluded for the following main reasons: wrong study design (n = 22) or absence of a control group (n = 15). Finally, 20 trials met all criteria and were included in the qualitative synthesis, all of which provided adequate data for meta-analysis (Fig. 1). PRISMA 2020 flowchart of the study selection process The study selection process was conducted independently by two reviewers. After duplicate removal, titles and abstracts were screened according to the predefined eligibility criteria. Potentially relevant studies were subsequently assessed through full-text review to determine final inclusion. Discrepancies between reviewers during both the screening and eligibility phases were resolved through discussion and consensus. When disagreement persisted, a third reviewer was consulted (Supplementary material 2). Evaluation of methodological quality Overall, the methodological quality of the included studies was moderate to high. Across the 20 trials, PEDro scores ranged from 5 to 8 points, with most studies demonstrating acceptable randomization procedures, baseline comparability, appropriate statistical comparisons, and adequate reporting of point estimates and variability. However, as expected in exercise-based trials, participant and therapist blinding were rarely feasible, and allocation concealment and intention-to-treat analysis were inconsistently reported. Despite these limitations, the majority of trials showed sufficient methodological rigor to support the reliability of their findings. The detailed PEDro ratings for each study are presented in Table 1. Table 1PEDro methodological quality assessment of the included studiesStudy1234567891011TotalAtılgan et al. [19]YYNYNNYYNYY6Boudreau et al. [20]YYYYNNYYYYY8Dejaco et al. [21]YYYYNNNYYYY7Eliason et al. [22]YYYYNNYYYYY8Gialanella et al. [23]YYNYNNYYNYY6Granviken et al. [24]YYYYNNYYYYY8Gutiérrez- Espinoza et al. [25]YYYYNNYYYYY8Gutiérrez- Espinoza et al. [26]YYYYNNYYYYY8Haider et al. [27]NYNYNNYNYYY5Heerspink et al. [28]YYNYNNYYNYY6Iğrek et al. [29]YYNYNNYYNYY6Ingwersen et al. [30]YYNYNNYYNYY6Juul-Kristensen et al. [31]YYYYNNYYYYY8Liaghat et al. [32]YYYNNYYYYYY8Littlewood et al. [33]YYYYNYYNNYY7Macías-Hernández et al. [34]YYYYNNYYNYY7Mintken et al. [35]YYYYNNYYYYY8Naunton et al. [36]YYNYYNYNNYY5Tahran et al. [37]YYNYNNYYNYY5Weekes et al. [38]YYYYNNYYNYY7(1) Eligibility criteria specified (not scored); (2) Random allocation; (3) Concealed allocation; (4) Baseline comparability; (5) Blinding of participants; (6) Blinding of therapists; (7) Blinding of assessors; (8) Adequate follow-up (> 85%); (9) Intention-to-treat analysis; (10) Between-group comparisons; 11. Point estimates and variabilityY Yes, N No PEDro methodological quality assessment of the included studies (1) Eligibility criteria specified (not scored); (2) Random allocation; (3) Concealed allocation; (4) Baseline comparability; (5) Blinding of participants; (6) Blinding of therapists; (7) Blinding of assessors; (8) Adequate follow-up (> 85%); (9) Intention-to-treat analysis; (10) Between-group comparisons; 11. Point estimates and variability Y Yes, N No Risk of bias assessment A risk of bias assessment was conducted using the Cochrane Risk of Bias 2 (RoB 2) tool. Overall, most studies showed either low risk of bias or some concerns, whereas only a limited number of trials were classified as high risk of bias. The most common source of bias was related to deviations from intended interventions, mainly due to the impossibility of blinding participants and therapists in exercise-based interventions. Some concerns were also frequently identified in the randomization process and in the selection of the reported results. The domains with the lowest risk were missing outcome data and measurement of the outcome, as several studies reported low attrition rates and used blinded assessors or validated outcome measures. However, a few studies presented high risk of bias because of inadequate randomization procedures, substantial attrition, lack of blinded outcome assessment, or important deviations from the assigned intervention. Overall, the methodological quality of the included randomized controlled trials was considered acceptable, although the inherent difficulties in blinding exercise interventions should be taken into account when interpreting the findings (Fig. 2). Risk of bias assessment of included studies using the cochrane risk of bias 2 (RoB 2) tool Certainty of evidence The certainty of evidence was assessed using the GRADE approach for the main outcomes included in this review. Overall, the certainty of evidence ranged from low to moderate. Evidence related to pain and functional outcomes was considered moderate, although downgraded because of methodological limitations identified in several studies, particularly regarding blinding procedures and deviations from intended interventions. Evidence for range of motion, muscle strength, and psychosocial outcomes was rated as low due to heterogeneity among interventions and outcome measurements, as well as imprecision associated with small sample sizes in several trials. No serious concerns regarding publication bias were identified (Table 2). Table 2GRADE assessment of the certainty of evidence for the main outcomesOutcomeCertainty of evidenceReasons for downgradingPainModerateRisk of bias due to lack of blinding and deviations from intended interventionsFunction/disabilityModerateRisk of bias and heterogeneity across interventionsRange of motionLowRisk of bias, inconsistency, and imprecisionMuscle strengthLowSmall sample sizes and heterogeneity in assessment methodsPsychosocial outcomesLowIndirectness and imprecisionAdverse eventsLowInconsistent reporting and imprecision GRADE assessment of the certainty of evidence for the main outcomes Characteristics of the studies A total of 20 randomized clinical trials met the inclusion criteria and were incorporated into this systematic review. These studies were published between 2015 and 2025 and were conducted across diverse clinical and research settings, including physiotherapy outpatient departments, rehabilitation centers, orthopedic clinics, and university-affiliated hospitals. Sample sizes varied considerably, ranging from small feasibility or pilot studies with fewer than 30 participants to larger trials enrolling more than 100 individuals. Regarding the population studied, most trials included adults diagnosed with subacromial pain syndrome, rotator cuff tendinopathy, partial-thickness rotator cuff tears, or degenerative rotator cuff disorders. Although the majority of samples represented middle-aged adults, several studies focused specifically on older populations, such as Gialanella et al. [23], who evaluated conservative exercise treatment in patients over 60 years with degenerative rotator cuff tears. The duration of symptoms across studies generally ranged from 3 months to several years, reflecting chronic or persistent shoulder pain conditions. The interventions varied substantially in type, intensity, and delivery. Most studies evaluated exercise-based programs, including strengthening, stretching, neuromuscular training, eccentric and concentric loading protocols, proprioceptive neuromuscular facilitation, scapular stabilization exercises, or home-based programs. Several trials examined the additive effect of adjunct treatments combined with exercise, such as manual therapy [27, 35], joint mobilization [22], EMG-biofeedback [31], psychomotor therapy [30], stretching of specific muscles [25], or relaxation techniques [38]. One trial compared high- vs. low-load volume exercise [36], while others focused on different exercise strategies (specific vs. general exercise, supervised vs. home-based, Pilates, or cycloergometer-based routines). Control groups also varied across trials and included usual physiotherapy management, exercise alone, placebo or sham interventions, stretching only, or no treatment in the case of Eliason et al. [22]. Only one included study [28] compared conservative exercise therapy with surgical repair, offering a unique perspective on treatment alternatives for degenerative tears. Outcome measures were heterogeneous but consistently addressed domains relevant to rotator cuff-related shoulder pain. Pain intensity was commonly assessed using the Visual Analog Scale (VAS) or Numeric Pain Rating Scale (NPRS). Shoulder function was measured with validated tools such as the Disabilities of the Arm, Shoulder and Hand questionnaire (DASH), the Shoulder Pain and Disability Index (SPADI), and the Constant-Murley Score (CMS). Other studies included strength assessments, range of motion, scapular kinematics, quality of life measures (e.g., SF-36), or exercise-induced pain trajectories. Intervention duration typically ranged from 4 to 16 weeks, with follow-up periods varying from immediately post-intervention to 6 months. Overall, the studies provide a broad overview of different exercise-based approaches for managing rotator cuff-related shoulder pain (Table 3). Table 3Characteristics of included studiesAuthorsStudy DesignPopulation / DiagnosisAgeSample Size (Exercise / Control)InterventionControl GroupOutcomes & MeasuresExercise modalityDurationExercise intensity / progressionAtılgan et al. [19]RCTShoulder pain52.46 ± 11.2130 (15/15)Clinical Pilates4 weeks5 sessions/week; 10 repetitions per exerciseConventional physiotherapyPain, ROM, functionBoudreau et al. [20]RCTRotator cuff tendinopathy50.2 ± 10.942 (21/21)GH adductor coactivation + RC exercises6 weeks3 sets of 10 repetitions; progression based on pain tolerance.Rotator cuff exercises alonePain, function, strengthCamargo et al. [21]RCTShoulder impingement syndrome42.65 ± 10.7346 (23/23)Stretch + strengthening (± manual therapy)4 weeks3 sets of 10 repetitions with progressive elastic resistance exercises.Stretching and strengthening without manual therapyPain, function, range of motionDejaco et al. [22]RCTRotator cuff tendinopathy50.2 ± 10.836 (20/16)Eccentric vs. conventional exercise26 weekseccentric exercises twice daily vs. conventional exercises once daily with progressive loading.Conventional exercise programPain, function, strengthEliason et al. [23]RCTSubacromial pain syndrome45.5 ± 8.378 (46/32)Guided exercises ± joint mobilization6 weeks20 supervised sessions (2/week) + home exercises twice daily; 3 sets of 10 repetitions with gradual progression.Guided exercise program without joint mobilizationPain, ROM, functionGialanella et al. [24]RCTDegenerative rotator cuff tear (older adults)79.6 ± 4.140 (20/20)Conservative care + cycloergometer6 months5 sessions/week; 30-min supervised exercise sessions plus home cycloergometer training (20 min twice daily) for 6 months.Conservative care alonePain, function, ROM, strengthGranviken et al. [25]RCTSubacromial impingement48.2 ± 9.818/21Home-based vs. supervised exercises6 weeksTwice daily; 3 sets of 30 repetitions with individualized progression.Supervised exercise / home-based exerciseFunction, painGutiérrez- Espinoza et al. [26]RCTSubacromial impingement syndrome45.7 ± 4.942 (26/26)Specific exercise program5 weeks2 sessions/week + daily home exercises; 8–10 repetitions with 5–10 s holds and graded progressionGeneral exercise programPain, functionGutiérrez- Espinoza et al. [27]RCTSubacromial pain syndrome44.5 ± 5.480(40/40)Exercise + PM stretching12 weeks3 sessions/week; 8–10 repetitions with 5–10 s holds, symptom-guided progression, plus daily home exercisesExercise program without PM stretchingPain, functionHaider et al. [28]RCTSubacromial pain49.80 ± 9.640 (20/20)Exercise + Maitland thoracic manipulation2 weeks3 sessions/week for 2 weeks; mobility and strengthening exercises with elastic resistance and progressive loading (1–4 kg).Exercise program alonePain, functionHeerspink et al. [29]RCTDegenerative rotator cuff tears60.8 ± 7.245 (25/20)Conventional exercises12 weeksrogressive rehabilitation with passive exercises during the first 6 weeks, followed by active and strengthening exercises after 3 months.Conservative exercise therapyPain, functionIğrek et al. [30]RCTSubacromial impingement syndrome47.6 ± 12.429 (15/14)PNF exercises16 weeks5 sessions/week; progression from ROM exercises to elastic-band strengthening from week 3 (3 sets × 10 repetitions).Conventional physiotherapyPain, function, ROMIngwersen et al. [31]RCTChronic shoulder pain50.4 ± 12.775 (37/38)Psychomotricity + exercises12 weeksEvery second day; 3 sets × 15–20 repetitions with progressive strengthening/stabilization exercisesExercise program without psychomotricityPain, function, quality of lifeJuul-Kristensen et al. [32]RCTSubacromial pain syndrome43.05 ±12.435 (17/18)Neuromuscular exercises ± EMG-biofeedback8 weeks1 supervised session/week + home exercises; scapular neuromuscular training with progressive motor-control tasks over 8 weeks.Neuromuscular exercise without EMG biofeedbackPain, functionLiaghat et al. [33]RCTHypermobility-related shoulder pain40.7 ± 13.264 (34/30)Shoulder strengthening16 weeks3 sessions/week; progressive shoulder strengthening from 50%–100% 10RM, progressing to 8RM over 16 weeks.Low-load strengthening / usual carePain, strengthLittlewood et al. [34]RCTRotator cuff tendinopathy53.8 ± 10.286 (42/44)Self-managed single exercise12 weeks2 sessions/day; 3 sets of 10–15 repetitions with progressive loading based on symptom responseUsual physiotherapy carePain, function, quality of lifeMacías-Hernández et al. [35]RCTPartial rotator cuff tear + moderate-severe shoulder pain54.5 ± 10.526 (12/14)Eccentric strengthening12 months5 sessions/week; 3 sets of 12 repetitions with progressive loading from 250 g to 1 kg according to tolerance.Conventional rehabilitationPain, strengthMintken et al. [36]RCTShoulder pain44.8 ± 12.9114 (57/57)CT manual therapy + exercise6 months2 sessions/week; 3 sets of 10 repetitions with 3-phase progressive strengthening using elastic resistanceExercise alonePain, functionNaunton et al. [37]RCTRotator cuff tendinopathy49.4 ± 12.231 (15/16)High vs. low load-volume exercise12 weeks2–4 sessions/week; 2–5 sets of 4–20 repetitions with periodized progressive loading (66–85% 1RM).Alternative exercise loading strategyPain, functionTahran et al. [38]RCTSubacromial impingement syndrome55 ± 9.745 (22/23)Modified posterior shoulder stretching4 weeksDaily stretching program; 1 set of 5 repetitions with 30-s holds over 4 weeks.Alternative stretching protocolROM, pain, functionWeekes et al. [39]RCTPost-arthroscopic rotator cuff repair60 ± 10119 (58/61)Relaxation exercises2 weeksRelaxation breathing exercises every 4 h during the first 5 postoperative days; progression from 10 to 20 breathing cyclesStandard postoperative rehabilitationPain, function, range of motionRCT randomized controlled trial, SIS subacromial impingement syndrome, RC rotator cuff, GH glenohumeral, ROM range of motion, PM pectoralis minor, PNF proprioceptive neuromuscular facilitation, CT cervicothoracic Characteristics of included studies 43.05 ± 12.4 RCT randomized controlled trial, SIS subacromial impingement syndrome, RC rotator cuff, GH glenohumeral, ROM range of motion, PM pectoralis minor, PNF proprioceptive neuromuscular facilitation, CT cervicothoracic Study results Across the 20 included randomized controlled trials, exercise-based interventions consistently produced meaningful improvements in pain, function, range of motion, and strength in individuals with rotator cuff–related shoulder pain. Although the specific exercise modalities varied, the overall trend across studies indicated that exercise was effective as a primary conservative intervention. Pain outcomes Reductions in pain intensity were reported in nearly all trials. Strengthening-based interventions, such as eccentric or conventional loading, significantly decreased pain in patients with rotator cuff tendinopathy [21, 33]. Studies evaluating specific exercise programs also demonstrated superior or comparable improvements in pain compared with general exercise [26]. The addition of glenohumeral adductor coactivation to a rotator cuff strengthening protocol further reduced pain compared with standard strengthening alone [20]. Adjunct techniques combined with exercise produced mixed but generally positive results. Thoracic spinal manipulation plus exercise led to greater pain reduction in the short term compared with exercise alone [27, 35], whereas joint mobilization added to guided exercise showed limited additional benefit [22]. Stretching-focused trials also showed pain improvements, including posterior shoulder stretching [37] and pectoralis minor stretching added to exercise [25]. High- vs. low-load strengthening programs produced similar reductions in pain [32, 36], suggesting that load volume may not be a critical determinant of pain relief. Functional outcomes Functional improvements were consistently observed across studies, measured through SPADI, DASH or Constant-Murley Score. Clinical Pilates demonstrated significant functional gains in patients with shoulder pain [19]. Specific vs. general exercise programs yielded functional improvements in both groups, though specific exercise sometimes produced greater benefits [26]. Neuromuscular exercise programs, with or without EMG-biofeedback, led to significant improvements in function, with modest additional benefit in the biofeedback group [31]. Psychomotricity combined with active shoulder exercise improved disability-related measures more than exercise alone [31]. In older adults with degenerative tears, functional gains were achieved following a conservative exercise plus cycloergometer program [23]. Postoperative outcomes also improved: relaxation exercises reduced pain-related functional limitations and analgesic needs after rotator cuff repair [38]. Range of Motion (ROM) Several trials demonstrated meaningful ROM gains. Stretching-based interventions showed the most pronounced effects, including modified posterior shoulder stretching [37] and stretching of the pectoralis minor [26]. Eccentric or concentric strengthening produced ROM improvements but generally less than stretching-focused protocols [34]. Joint mobilization adjuncts, particularly when paired with guided exercise, yielded moderate ROM improvements [23]. Strength outcomes Strength gains were reported in trials involving structured strengthening programs. Eccentric strengthening improved shoulder strength more effectively than concentric exercise [34]. Programs focused on rotator cuff loading demonstrated increased isometric and dynamic strength [33]. The addition of adductor coactivation enhanced rotator cuff muscle recruitment and strength outcomes compared with standard strengthening [20]. High- and low-load protocols resulted in similar strength gains, reinforcing the adaptability of the rotator cuff to various load volumes [36]. Additional outcomes Some studies evaluated complementary domains. Psychomotor therapy yielded benefits in body awareness and movement coordination [30]. High-load and low-load interventions were associated with distinct patterns of exercise-induced pain trajectories [32]. Post-surgical trials showed that relaxation exercises reduced opioid use and improved postoperative pain tolerance [38]. Meta-analysis A total of 20 studies were included in the meta-analysis to examine the effects of physical interventions on the rotator cuff in middle-aged and older adults. The heterogeneity analysis revealed a Q value of 194.050 with 45 degrees of freedom, indicating substantial heterogeneity among the included studies. Consistent with this finding, the I² value was 76.81%, suggesting that approximately 77% of the observed variability was attributable to true differences in effect sizes rather than sampling error. Furthermore, the Tau² (0.263) and Tau (0.513) values also confirmed considerable dispersion in the true effects across studies. Given this high heterogeneity, a random-effects model was applied. The overall effect size was − 0.103 (95% CI: -0.286 to -0.060), which was statistically significant (p = 0.013), indicating a small beneficial effect of physical interventions on rotator cuff–related symptoms in this population. For this exploratory global synthesis, outcomes from different clinical domains (e.g., pain, function, range of motion, and strength) were standardized using Hedges’ g to enable comparisons across studies employing different assessment instruments and scales. However, because these outcomes represent distinct clinical constructs, separate subgroup meta-analyses were subsequently conducted for each outcome category to provide a more clinically meaningful interpretation of the findings. Although the overall pooled effect size was small, several individual studies reported statistically significant and clinically relevant effects (Fig. 3). Forest plot showing the effect of exercise interventions on pain and physical function in middle-aged and older adults Subgroup analysis Pain A subgroup analysis was performed to evaluate the effect of the interventions on pain, using a total of 19 effect sizes. The analysis revealed a pooled effect size of g = -0.271 under the random-effects model, with a 95% confidence interval of [-0.393, -0.148]. This result was statistically significant (p < 0.001) and suggests a moderate to large effect in favor of the pain-reducing interventions. Heterogeneity among the studies was substancial, as demonstrated by the Q-value (Q = 77.488) with 18 degrees of freedom (p < 0.001) and an I² value of 76.77%, suggesting that most of the observed variability across studies was attributable to true differences rather than chance. Furthermore, the tau-square value (τ² = 0.251) indicated considerable dispersion in effect sizes among the included studies. The forest plot (Fig. 4) shows the variability in the effect sizes of the individual studies and the combined effect, which is consistent with the magnitude of the effect observed in the analysis. Forest plot showing the effect sizes of the individual studies and the combined effect on the pain variable Regarding publication bias, the funnel analysis (Fig. 5) showed some asymmetry, but the statistical tests, both Begg’s test (p = 0.046) and Egger’s test (p = 0.029), were significant, suggesting that publication bias is a major factor in the observed results. As for the Orwin fail-safe N analysis, no further studies would be needed to invalidate the results. Funnel plot that evaluates asymmetry and publication bias in the pain variable Function For the variable “function,” the meta-analysis results revealed a moderate effect size (g = 0.159) with a 95% confidence interval ranging from 0.037 to 0.281. This effect size is statistically significant, as the p-value is less than 0.05 (p = 0.010). Furthermore, the heterogeneity analysis indicated moderate variability between studies, with a Q-value of 64.724 (p < 0.001) and an I² of 72.19%, suggesting that approximately 72% of the observed variability between studies is due to genuine differences rather than random errors. The Tau-squared value was 0.195, supporting this variability. The random-effects model was the most appropriate for this analysis due to the considerable heterogeneity observed. The florest plot (Fig. 6) shows no apparent publication bias, although the shape of the plot should be considered. Forest plot showing the effect sizes of the individual studies and the combined effect on the function variable Publication bias analysis using Egger’s test revealed significant evidence of bias (p = 0.029), and Begg and Mazumdar’s analysis showed significant correlation (p = 0.048). Finally, Duval and Tweedie’s trim and fill analysis indicated that no adjustments for missing studies were necessary, as no significant missing studies were detected in the calculations (Fig. 7). Funnel plot that evaluates asymmetry and publication bias in the function variable ROM The meta-analysis results revealed a standardized overall effect size (g) of 0.454, indicating a moderate effect of physical interventions on range of motion. This effect was statistically significant (p = 0.028), as the 95% confidence interval [0.049, 0.859] does not include zero, suggesting a true effect of interventions to improve range of motion. To assess heterogeneity among the studies, a Q test was performed, yielding a Q value of 7.933 with 4 degrees of freedom. The results indicated moderate heterogeneity among the included studies. This finding was supported by the I² value of 49.58%, suggesting that approximately 50% of the observed variability across studies was attributable to true differences in effect sizes rather than chance. Furthermore, the tau-square value (τ² = 0.106) and Tau value (τ = 0.151) indicated moderate dispersion in the true effect sizes among the studies. Therefore, a random-effects model was considered appropriate for this analysis, as it accounts for variability arising from differences in study design, participant characteristics, and intervention protocols (Fig. 8). Forest plot that evaluates asymmetry and publication bias in the ROM variable Although the Egger test (p = 0.037) suggests potential publication bias, this should be interpreted with caution due to the relatively small number of included studies and the high heterogeneity among them (Fig. 9). Funnel plot that evaluates asymmetry and publication bias in the ROM variable Strength The analysis of the “strength” variable showed a standardized effect size (g) of 0.544 under the random-effects model, indicating a moderate positive effect of physical interventions on strength outcomes. This effect was statistically significant (p = 0.020), with a 95% confidence interval ranging from 0.086 to 1.001, which did not include zero. These findings suggest that physical interventions may contribute to improvements in strength among middle-aged and older adults with rotator cuff–related conditions. The heterogeneity analysis yielded a Q value of 12.544 (p = 0.248), indicating no statistically significant heterogeneity among the included studies. However, the I² statistic of 68% suggested substantial variability between studies. In addition, the tau-square value (τ² = 0.035) and Tau value (τ = 0.188) indicated some variability in the true effect sizes across studies. Therefore, a random-effects model was considered the most appropriate approach for this analysis. (Fig. 10). Forest plot that evaluates asymmetry and publication bias in the strength variable The funnel plot suggests a symmetrical distribution, which also supports the absence of publication bias. Regarding publication bias, the Egger test (p = 0.071) did not indicate significant publication bias, and the funnel plot showed a relatively symmetrical distribution (Fig. 11). Funnel plot that evaluates asymmetry and publication bias in the strength variable Discussion The main objective of this meta-analysis was to synthesize and quantify the available scientific evidence on the effects of physical interventions in middle-aged and older adults with rotator cuff pain and dysfunction. Through an analysis of 20 studies, the magnitude of the impact of different exercise modalities on key variables, such as pain, function, range of motion, and strength, was evaluated in order to provide relevant information for future interventions and research in this area. The results suggest that physical interventions have a small but statistically significant positive effect on improving rotator cuff-related symptoms, although considerable heterogeneity was observed among studies. The overall analysis showed an effect size of -0.103 (p = 0.013), indicating a modest improvement in general symptoms, with more consistent effects observed for pain and range of motion outcomes. These findings should also be interpreted in the context of current clinical practice guidelines, which generally recommend exercise as a first-line conservative treatment for rotator cuff-related shoulder pain. However, many previous systematic reviews and clinical recommendations have evaluated exercise as part of multimodal rehabilitation approaches that combine manual therapy, electrotherapy, injections, or pharmacological interventions. In contrast, the present review focused specifically on exercise-based interventions in which exercise remained the primary therapeutic component, even when some studies incorporated adjunct techniques such as manual therapy or stretching. This approach allowed a more focused evaluation of the contribution of exercise to pain, function, range of motion, and strength outcomes in middle-aged and older adults with rotator cuff-related shoulder pain [40, 41]. However, the substantial heterogeneity observed across studies (I² = 76.81%) suggests that an important proportion of the variability between studies may be attributable to genuine differences in interventions, exercise protocols, outcome measures, and participant characteristics rather than to sampling error alone.Pain associated with rotator cuff injuries is a major symptom in middle-aged and older adults, and its management represents a central objective of rehabilitation programs [42, 43]. In the present meta-analysis, physical interventions demonstrated a statistically significant beneficial effect on pain reduction, with a pooled effect size of g = -0.271 (p < 0.001). Although the magnitude of the pooled effect was small-to-moderate, these findings support the effectiveness of exercise-based interventions for reducing pain in this population. Our results are consistent with those reported by Page et al. [42], who also observed significant improvements in pain following physical rehabilitation interventions, particularly strengthening and stretching exercises in individuals with rotator cuff-related shoulder disorders. However, many previous reviews primarily evaluated broader multimodal rehabilitation approaches combining exercise with manual therapy and other co-interventions, making it more difficult to isolate the specific contribution of exercise itself. In contrast, exercise remained the primary therapeutic component across the studies included in the present review, allowing a more focused evaluation of exercise-based rehabilitation strategies. Furthermore, our review incorporated a broad range of exercise modalities, including strengthening exercises, scapular stabilization, stretching, neuromuscular exercise, and progressive loading programs, providing a more comprehensive overview of how different exercise approaches may contribute to pain reduction. This finding is also consistent with previous systematic reviews, such as that by Littlewood et al. [44], which reported favorable evidence supporting exercise interventions for pain reduction in rotator cuff tendinopathies. Similarly, Steuri et al. [45] concluded that structured exercise programs, including both high-load and low-load strengthening protocols, are effective in reducing pain and improving shoulder-related outcomes in this clinical population. The relevance of exercise interventions may be particularly important in middle-aged and older adults due to the age-related pathophysiological changes commonly associated with rotator cuff disorders. Aging is associated with progressive tendon degeneration, reduced vascularity, altered collagen organization, diminished tendon load tolerance, sarcopenia, and declines in neuromuscular control, all of which may contribute to persistent pain and functional impairment. Therapeutic exercise may therefore exert beneficial effects not only through pain modulation mechanisms, but also by improving tendon load capacity, muscular performance, neuromuscular coordination, and movement efficiency during daily activities [46–49]. Consequently, although previous literature already supported physical exercise as a valid conservative intervention, the present review further reinforces the evidence that exercise-based rehabilitation should be considered a first-line treatment approach for middle-aged and older adults with rotator cuff-related shoulder pain. Functional improvement is a key rehabilitation goal in patients with rotator cuff injuries, as shoulder dysfunction significantly affects daily activities and quality of life [50]. In the present review, exercise-based interventions generally demonstrated favorable effects on functional outcomes, supporting the role of therapeutic exercise in improving shoulder-related disability and functional performance in middle-aged and older adults. Functional outcomes were commonly assessed using validated instruments such as the SPADI, DASH, and Constant-Murley scales, which are widely recognized in the literature for evaluating disability and shoulder function in patients with rotator cuff-related disorders. The functional improvements observed across studies may be explained by the ability of exercise interventions to enhance muscular performance, scapulohumeral coordination, joint mobility, and neuromuscular control during daily activities. A broad range of exercise modalities was represented in the included studies, including strengthening exercises, scapular stabilization programs, neuromuscular exercise, and clinical Pilates. In particular, Pilates-based interventions demonstrated promising functional benefits, potentially due to their emphasis on postural control, movement coordination, breathing regulation, and functional movement patterns. These findings are consistent with those reported by Mueller et al. [51], who observed significant improvements in the functional capacity of older women with shoulder pain following Pilates-based exercise interventions. Furthermore, comparative analyses between functional and conventional exercise programs generally revealed improvements in both groups, although additional benefits were sometimes observed with more specific exercise approaches. This finding is consistent with previous literature suggesting that targeted strength and conditioning programs may provide superior functional outcomes compared with more generalized rehabilitation interventions. Hlaing et al. [52] similarly reported that specific exercise programs improved muscle characteristics and functionality in patients with nonspecific subacute low back pain. Neuromuscular exercise interventions also appeared to contribute positively to functional recovery, particularly when combined with electromyographic biofeedback strategies aimed at optimizing muscle activation patterns. Similar findings have been reported by Zhong et al. [48, 53], who demonstrated that neuromuscular training modalities, including biofeedback, can improve muscle activation and exercise performance in individuals with musculoskeletal disorders. Collectively, these findings reinforce the importance of individualized and progressively dosed exercise interventions for improving shoulder-related function in middle-aged and older adults with rotator cuff-related disorders. Improving shoulder range of motion (ROM) is an essential goal in rotator cuff rehabilitation, as mobility limitations significantly interfere with daily activities and quality of life in middle-aged and older adults [54]. In the present review, exercise-based interventions demonstrated favorable effects on ROM outcomes, reinforcing the therapeutic value of physical interventions targeted at this population. Improvements in ROM may be associated with reductions in pain, enhanced neuromuscular control, improved muscular flexibility, and increased tolerance to tendon loading achieved through progressive exercise interventions. These findings are supported by recent studies such as that by Chepeha et al. [55], who demonstrated in a prospective cohort study that a progressive, criteria-based therapeutic exercise program in patients with rotator cuff-related shoulder pain produced statistically significant improvements in active ROM, particularly in abduction and external rotation, as early as 6 weeks and sustained up to 26 weeks. Similarly, the systematic review by Ankar et al. [56], which compared different dynamic stretching interventions for rotator cuff injuries, concluded that more active exercise protocols resulted in greater improvements in shoulder mobility. Rotator cuff strength is crucial for dynamic shoulder stability, as it plays a fundamental role in movement control and upper-limb function during everyday activities [57]. Although exercise interventions generally improve pain and ROM, restoring muscular strength in patients with rotator cuff disorders may be more challenging due to tendon degeneration, muscle atrophy, pain-related inhibition, and age-related declines in muscular performance [58]. In the present meta-analysis, exercise interventions demonstrated a statistically significant positive effect on strength outcomes, with an effect size of g = 0.544 (p = 0.020), indicating a moderate beneficial effect of exercise on muscular strength. These findings are consistent with the review by Vila-Diegueze et al. [59], which highlighted that progressive loading exercises promote the muscular and neuromuscular adaptations necessary to improve strength in individuals with rotator cuff tendinopathy. However, previous literature also suggests that strength gains often require longer intervention durations and appropriate progression strategies, particularly in older adults and individuals with chronic tendon pathology. This may partly explain the variability observed across studies included in the present review. The moderate improvement in strength observed following exercise interventions may be explained by several factors. Although physical exercise demonstrated a statistically significant positive effect on strength outcomes, muscle atrophy and tendon degeneration commonly associated with rotator cuff disorders may limit the magnitude and speed of strength recovery, particularly in individuals with more severe or chronic conditions. Furthermore, the design of exercise programs, session frequency, progression criteria, and training intensity are all likely to influence strength-related outcomes. In this regard, our review highlights the importance of progressive and individualized exercise programs that appropriately manipulate load, intensity, and exercise variability to optimize strength gains. These considerations are especially relevant in older adults, in whom reduced tissue adaptability and recovery capacity may increase sensitivity to excessive loading. Consequently, exercise prescription in this population should emphasize gradual progression, symptom monitoring, individualized load management, and long-term adherence strategies to optimize clinical outcomes while minimizing symptom exacerbation [60]. The observed improvements in strength are clinically relevant, as restoring muscular capacity is fundamental for maintaining dynamic shoulder stability, improving functional performance, preventing recurrent symptoms, and supporting independence during daily activities. Therefore, exercise-based rehabilitation should remain a central component of conservative management for rotator cuff-related shoulder disorders, with particular emphasis on individualized progression and long-term therapeutic adherence. It is important to note that many of the studies included in this meta-analysis presented methodological limitations, such as lack of blinding in some trials, heterogeneous inclusion criteria, and relatively small sample sizes. These limitations may have affected the internal validity and generalizability of the findings. Furthermore, the considerable heterogeneity observed across intervention protocols, including variability in exercise frequency, intensity, duration, progression criteria, and outcome assessment methods, may limit the ability to establish definitive conclusions regarding the optimal exercise approach for rotator cuff-related shoulder disorders. In addition, although several exercise modalities were identified across studies (e.g., strengthening, stabilization, stretching, neuromuscular training, Pilates, and eccentric exercise), the variability in intervention protocols, exercise dosage, comparator groups, and outcome reporting limited the feasibility of conducting robust quantitative subgroup meta-analyses according to exercise type. Consequently, exercise modalities were analyzed primarily in a descriptive manner in the present review. Additionally, the pooled global analysis combined outcomes from distinct clinical domains, including pain, function, range of motion, and strength, which may have diluted the overall effect size due to differences in responsiveness across outcome measures. Another important limitation is that some included studies combined exercise with adjunctive interventions such as manual therapy, stretching, or thoracic manipulation. Although exercise remained the primary therapeutic component and these studies met the eligibility criteria, the presence of co-interventions may have influenced the observed effects and limits the ability to attribute outcomes exclusively to exercise alone. Given these limitations, future research should prioritize the standardization of exercise protocols and reporting methods in order to improve comparability across studies. Future rehabilitation trials should also adopt more standardized reporting of exercise dosage variables, including intensity, load progression criteria, supervision level, adherence monitoring, and long-term follow-up strategies. Furthermore, multicenter studies with larger and more diverse samples are needed, together with the incorporation of objective assessment tools such as electromyography and biomechanical movement analysis to better evaluate intervention effects on strength, mobility, and functional performance. Future studies should also consider stratifying participants according to age, symptom duration, tendon pathology severity, and comorbidity profiles in order to identify which exercise approaches may be most effective for specific clinical subgroups of middle-aged and older adults. The use of standardized core outcome sets and clinically meaningful follow-up periods would further improve comparability across studies and strengthen the translation of evidence into rehabilitation practice. Additionally, future research could explore the effects of combining scapulohumeral stabilization exercises with multimodal rehabilitation strategies, such as manual therapy or electrotherapy, to determine whether combined approaches provide superior clinical outcomes. Finally, variability in diagnostic terminology across included studies represents an additional limitation. Some trials used broader clinical terms such as “shoulder pain” or “subacromial pain syndrome” rather than strictly imaging-confirmed rotator cuff diagnoses. Although these conditions are frequently considered part of the spectrum of rotator cuff-related shoulder disorders in clinical practice, this variability may have introduced additional heterogeneity into the pooled analyses. Conclusions Current evidence supports the effectiveness of exercise-based interventions for the rehabilitation of rotator cuff-related shoulder pain in middle-aged and older adults. The present review demonstrates that structured therapeutic exercise can produce statistically significant improvements in pain, function, range of motion, and muscular strength in this age-specific population, although the magnitude of the pooled global effect was small and considerable heterogeneity was observed across studies. Outcome-specific analyses nevertheless revealed clinically meaningful benefits, particularly for pain reduction, shoulder mobility, and strength outcomes. These findings reinforce the role of exercise as a first-line conservative treatment strategy for middle-aged and older adults with rotator cuff-related shoulder disorders. Importantly, this review expands previous evidence by specifically focusing on middle-aged and older adults, a population characterized by age-related tendon degeneration, reduced muscle capacity, sarcopenia, and altered neuromuscular function, all of which may influence rehabilitation responses and exercise tolerance. The findings therefore provide clinically relevant information to support exercise prescription and rehabilitation planning in this demographic group. However, the results should be interpreted with caution due to the considerable heterogeneity observed across studies regarding exercise modalities, intervention duration, intensity, progression criteria, co-interventions, outcome measures, and participant characteristics. In addition, methodological limitations such as relatively small sample sizes, variability in diagnostic terminology, and inconsistent reporting of exercise dosage, supervision, and adherence may have influenced the pooled results and limited direct comparisons between intervention protocols. Furthermore, the global pooled analysis combined outcomes from distinct clinical domains, which may have diluted the overall effect size. Future research should prioritize the development of more standardized and age-specific rehabilitation protocols, including clearer reporting of exercise intensity, progression strategies, supervision, adherence monitoring, and long-term follow-up outcomes. Additional high-quality multicenter randomized controlled trials are needed to determine which exercise approaches are most effective for specific clinical presentations and subgroups of middle-aged and older adults with rotator cuff-related shoulder pain. Supplementary Information Supplementary Material 1. Supplementary Material 2. Acknowledgments Institutional review board statement Not applicable. Informed consent statement Not applicable. Abbreviations Randomized controlled trial Subacromial impingement syndrome Rotator cuff Glenohumeral Range of motion Pectoralis minor Proprioceptive neuromuscular facilitation Cervicothoracic Shoulder pain and disability index Disabilities of the arm, shoulder, and hand questionnaire Constant-murley score Electromyography Authors’ contributions Conceptualization, P.V.-R., J.G.-I. and M.C.-O.; methodology, M.A.M.-A.; F.H.-C. and J.G.-I.; formal analysis, Y.C.-C. AND M.A.M.-A.; writing-original draft preparation, M.C.-O. and A.A.-A.; writing-review and editing, P.V.-R. and M.A.M.-A.; supervision, A.A.-A.; F.H.-C. and Y.C.-C. All authors have read and agreed to the published version of the manuscript. Funding This research received no external funding. Data availability Not applicable. Declarations Ethics approval and consent to participate Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1. Supplementary Material 2. Data Availability Statement Not applicable.