How Pragmatic Are Sarcopenia Intervention Studies? A Systematic Review

Authors: Sophie Van Heden, Zoubayda Baoubbou, Dolores Sanchez‐Rodriguez, Yoke Mun Chan, Charlotte Beaudart

Abstract

Background

Sarcopenia is an age‐related muscle disease often accompanied by comorbidities, mobility issues and cognitive decline, which can limit treatment adherence in older adults. Owing to the reversible nature of sarcopenia, there has been a growing number of randomized controlled trials conducted in recent years. Yet, many randomized controlled trials (RCTs) are conducted under ideal conditions (explanatory trials), limiting their real‐world applicability. In contrast, pragmatic trials aim to better reflect the complexities of clinical practice.

Objective

This study is aimed at assessing the level of pragmatism in current sarcopenia RCTs and identifying design gaps to further improve the clinical relevance and feasibility of future trials in the real world.

Methods

A systematic review was conducted on MEDLINE (via Ovid), Embase and Cochrane Central Register of Controlled Trials (PRISMA guidelines; PROSPERO: CRD42024571027). Eligible studies included RCTs on sarcopenia treatment using a consensus definition and published until March 2024. The PRECIS‐2 tool was used to assess the level of pragmatism of these RCTs across nine standard domains (eligibility, recruitment, setting, organization, flexibility of delivery, flexibility of adherence, follow‐up, primary outcome and primary analysis), with an additional ‘control’ domain. The total PRECIS‐2 score was calculated, and subgroup analyses were conducted by intervention type, geographical location, sample size, study duration and sarcopenia definition. A higher PRECIS‐2 score indicates greater trial pragmatism. The risk of bias was assessed using the Cochrane Risk of Bias 2.0 tool.

Results

Of the 3985 references reviewed, 54 RCTs met the inclusion criteria. The mean PRECIS‐2 score across its 10 domains was 2.93 (SD 1.30), reflecting a balance of explanatory and pragmatic characteristics. Organization, recruitment and primary outcome were identified as the most pragmatic domains, whereas eligibility, adherence and follow‐up were the most explanatory. Subgroup analyses revealed that geographical location and sarcopenia definitions impacted significantly the overall PRECIS‐2 score. More precisely, studies conducted in Asia achieved higher pragmatism scores, with significant differences in setting (p = 0.029), follow‐up (p = 0.014) and control (p = 0.042) domains. Studies using Asian sarcopenia criteria (e.g., AWGS) were also more pragmatic, particularly in the eligibility (p = 0.031) and control (p = 0.005) domains.

Conclusion

This systematic review reveals a persistent gap between explanatory and pragmatic designs in sarcopenia trials. Despite growing research, few studies reflect real‐world conditions. Key domains like eligibility, adherence and follow‐up remain overly controlled. Greater pragmatism is needed to ensure future trials yield evidence that is both robust and clinically applicable.

Introduction

Sarcopenia is a progressive disease characterized by the loss of muscle strength and mass, with physical performance used to assess its severity, as defined by the European Working Group on Sarcopenia in Older People (EWGSOP2) [1, 2]. Since 2016, it has been officially recognized as a distinct clinical entity, with the allocation of a specific ICD‐10‐CM code [3, 4].

The Global Leadership Initiative on Sarcopenia (GLIS) was recently established to harmonize existing definitions and develop a universally accepted standard [5]. This effort is particularly critical given that sarcopenia affects more than 10% of adults aged 65 years and older [6, 7], and is associated with numerous health consequences, including decreased physical performance and mobility, impaired ability to perform daily living activities, reduced quality of life, increased risk of falls and fractures, hospitalizations, greater likelihood of admission to long‐term care facilities and higher mortality rates [8, 9].

Sarcopenia is considered a potentially reversible disease [10]. This underscores the importance of interventional research aimed at identifying effective therapeutic strategies. To date, recommended treatments primarily focus on exercise, both endurance and resistance training [11, 12] and nutritional interventions, such as supplementation with vitamin D, protein, polyunsaturated fatty acids, antioxidants and adequate caloric intake [12]. The combination of physical activity and nutritional strategies currently represents the most effective approach [10, 13, 14]. No pharmacological treatment is yet approved for this indication, although several promising molecules, including apelin and irisin, are under investigation [15, 16].

Clinical trials investigating sarcopenia face numerous methodological and practical challenges. Older adults, and particularly older adults with sarcopenia, frequently present with multiple comorbidities [17, 18], geriatric syndromes and functional [8, 9], or cognitive impairments [19, 20, 21, 22] that may limit participation and adherence to highly controlled trial protocols. As sarcopenia is frequently accompanied by multimorbidities, strictly explanatory randomized controlled trials (RCTs) may fail to capture the complex clinical realities of these patients. Consequently, such trials often exclude the populations most affected, thereby limiting the external validity and real‐world applicability of their findings [23]. These challenges underscore the importance of pragmatic trial designs, which better reflect routine clinical practice and ensure that study results are directly relevant to those most affected by sarcopenia [23].

While pragmatic clinical trials are specifically designed to evaluate the effectiveness of interventions under real‐world conditions by accommodating clinical diversity, flexible protocols and outcomes relevant to both patients and healthcare providers [23, 24], much of the current evidence on sarcopenia still stems from highly controlled trial protocols. This limits the generalizability and practical value of research findings for routine clinical practice. To date, there is an information gap regarding the level of pragmatism along the pragmatic–explanatory continuum in clinical trials on sarcopenia. Bridging this gap is essential for enhancing the relevance and applicability of research findings to everyday clinical practice and for developing tailored therapeutic interventions that address the unmet needs of older adults with sarcopenia who face complex, real‐world health challenges.

The aim of this study is to assess the level of pragmatism within the pragmatic–explanatory continuum in published RCTs investigating interventions for managing sarcopenia in older adults. This approach seeks to identify methodological gaps and provide recommendations to enhance the clinical relevance and feasibility of future trials in real‐world settings.

Methods

Generality

This systematic literature review was carried out in accordance with the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) [25], ensuring methodological rigour and transparency. The completed PRISMA checklist is available in appendix (Table S1). The study protocol has been registered in the international PROSPERO database (registration CRD42024571027).

To ensure transparency and facilitate further exploration, all relevant documents and supplementary materials have been made publicly available on the Open Science Framework (OSF) platform (https://osf.io/bx429/).

Search Strategy and Study Selection

A comprehensive literature search was performed in March 2024 across Medline (via Ovid), Embase and the Cochrane Central Register of Controlled Trials (CENTRAL), using a predefined search strategy detailed in appendix (Table S2), to identify all RCTs specific to sarcopenia. Additional sources included manual screening of reference lists from relevant publications, citation tracking via Web of Science and consultations with domain experts. Previous systematic reviews and meta‐analyses on sarcopenia intervention studies were also reviewed to identify further eligible studies.

All studies retrieved from electronic databases and manual searches were imported into Covidence, a web‐based systematic review management platform designed to facilitate efficient screening, data extraction and other literature reviews. The study criteria for this systematic review are presented in Table 1.

All retrieved records were independently screened by reviewers (CB, DSR and YMC) based on titles and abstracts, followed by full‐text assessments to determine eligibility. Each study required evaluation by at least two reviewers, and any disagreements were resolved through discussion to reach consensus. The study selection process is outlined in the PRISMA flow diagram (Figure 1), with reasons for exclusion clearly documented and reported.

Data Extraction

The extracted data were recorded in a standardized Excel file that had been pretested on a sample of five studies. Data extraction was performed independently by two reviewers (SVH and ZB) to ensure accuracy and minimize bias. The extracted data included general study information, such as study design, study sample, study duration, geographical location, inclusion criteria, exclusion criteria, recruitment details, organization, type of intervention and arms, details of intervention group and control group, adherence, follow‐up, type of analysis (per protocol or intention‐to‐treat) and primary outcome.

If a RCT was included in several studies, only the first published study was considered in the statistical analysis (since the PRECIS‐2 pragmatic domains are identical for a single RCT) to avoid over‐representation of a single trial.

Pragmatism Assessment

Pragmatism was assessed using the PRagmatic Explanatory Continuum Indicator Summary (PRECIS)‐2 tool [27, 28]. It was developed to help trialists consider where their studies lie on the pragmatic/explanatory continuum during the design phase. The tool can also be used in a retrospective approach to address the level of pragmatism of published clinical trials [29]. PRECIS‐2 includes nine eligibility, recruitment, setting, organization, flexibility in delivery, flexibility in adherence, follow‐up, primary outcome and primary analysis. In addition to the original PRECIS‐2 domains, this study incorporated an extra domain assessing the design of control groups. Although this is not formally part of the PRECIS‐2 tool, Zwarenstein et al. [29], who were involved in the refinement of the original PRECIS into PRECIS‐2 [30], have suggested the inclusion of a control‐specific domain to better assess whether control conditions align with pragmatic principles, particularly in retrospective evaluations of clinical trials. This adaptation was made to enhance the tool's ability to assess pragmatism in the context of sarcopenia trials.

Pragmatism assessment was performed independently by two reviewers (SVH and ZB). Before starting the full evaluation, both reviewers jointly assessed two sample trials to calibrate their understanding of the tool and refine the application of the scoring criteria. Any disagreements were resolved through consensus, with the option of consulting a third party (CB) if necessary to reach a final decision. Each domain scores from 1 to 5, and a higher PRECIS‐2 score indicates greater pragmatism. A scoring table (Table 2) was developed based on the PRECIS‐2 guidelines [31], existing literature [32, 33] and the critical evaluation of the research team. This scoring methodology was applied to each study.

Risk of Bias Assessment

The quality of individual RCTs was evaluated using the Cochrane Risk of Bias 2.0 tool, applied independently by two reviewers (YMC and SVH). Any disagreements were resolved through consensus, with the option of consulting a third party (DSR) if necessary to reach a final decision [34].

Statistical Analysis

Descriptive statistics were used to summarize the PRECIS‐2 scores across domains. Specifically, the mean and standard deviation (SD) were calculated for each domain to provide an overall view of trial pragmatism. To enhance interpretability, graphical representations of PRECIS‐2 profiles were created using radar plots (‘PRECIS‐2 wheels’), allowing for a visual assessment of the degree of pragmatism across RCTs. Subgroup analyses were conducted to explore variations in PRECIS‐2 scores according to key study characteristics. Comparisons were made based on intervention type (exercise, nutrition, combined intervention, pharmacological treatments and other unclassified interventions), geographical location (continents), sample size (< 75 people, > 75 people—median of included RCTs), study duration (≤ 12 weeks, > 12 weeks—median of included RCTs), and definitions of sarcopenia (EWGSOP, AWGS, and other). Inter‐rater reliability of PRECIS‐2 scoring was assessed using intraclass correlation coefficients (ICC) [35].

Statistical comparisons between groups were performed using Student's t‐test for pairwise comparisons and one‐way ANOVA for comparisons involving more than two groups. Two sensitivity analyses were carried out to assess the quality of our one excluding the ‘control’ domain, and the other excluding studies with a high risk of bias. All statistical analyses were performed using R software. The data set and study codebook are available on Open Science Framework (https://osf.io/bx429/).

Results

After removing duplicates, a total of 2743 records were screened based on title and abstract. Of these, 147 articles underwent full‐text evaluation to assess eligibility. Ultimately, 54 RCTs met the inclusion criteria and were retained for the systematic review (Figure 1, PRISMA flowchart). Detailed reasons for the exclusion of the remaining studies are provided in the supplementary material (Table S3).

Studies Description

Table 3 presents the characteristics of the 54 included RCTs. These RCTs, published between 2012 and 2024, utilized various definitions of sarcopenia, with EWGSOP 2010 being the most frequently applied (20 RCTs, 37.0%), followed by AWGS 2014 (19 RCTs, 35.2%). Of the RCTs included in the review, 16 (29.6%) had fewer than 50 participants, 21 (38.9%) had between 51 and 100 participants and 17 (31.5%) had more than 100 participants. Most studies lasted between 10 and 20 weeks (24 RCTs, 44.4%), followed by studies lasting between 20 and 30 weeks (14 RCTs, 25.9%). In terms of interventions, exercise‐based interventions were the most common (19 RCTs, 35.2%), followed by a combination of both exercise and nutrition approaches (17 RCTs, 31.5%), nutritional interventions (11 RCTs, 20.4%), pharmacological interventions (four RCTs, 7.4%) and other types of interventions (three RCTs, 5.6%). Geographically, the majority of RCTs were conducted in Asia (31 RCTs, 57.4%), followed by Europe (13 RCTs, 24.1%) and America (seven RCTs, 13.0%). The remaining three RCTs (5.6%) were from other regions, including Oceania and various countries. Most of the recruitment was community‐based (36 RCTs, 66.7%), and the majority of the studies were multicentre (33 RCTs, 61.1%). Regarding funding, 22 RCTs (40.7%) received academic funding, and 19 RCTs (35.2%) received industrial funding. A detailed description of each included RCT is available in the supplementary material (Table S4).

The risk of bias was evaluated using Cochrane Risk of Bias 2.0. For analyses according to the intention‐to‐treat approach, 68.9% of studies were at low risk of bias, 15.6% raised some concerns and 15.6% were at high risk of bias (seven studies). The domains most frequently associated with a high risk of bias were missing outcome data and selection of the reported results (Table S5). On the other hand, for analyses conducted according to the per‐protocol approach, 55.6% of studies were judged to be at low risk of bias, while 44.4% raised some concerns. In these cases, the most frequently implicated domains were missing outcome data and selection of the reported results (Table S5). Studies judged at high risk of bias in the ‘missing outcome data’ domain typically had, for example, high dropout rates, with limited information on the nature of the dropouts. Studies judged at high risk of bias in the ‘selection of the reported results’ domain showed uncertainty regarding the designation of primary outcomes and possible selective reporting of results.

Results of PRECIS‐2

Overall Results

The average PRECIS‐2 score for all selected RCTs across all 10 domains was 2.93 (SD 1.30), indicating an overall balance between the explanatory and pragmatic levels of these studies. The study with the highest score had a score of 4.00 (an exercise‐based intervention conducted in Singapore, published in 2019) [36], and the study with the lowest score had a score of 2.13 (international study with nutritional intervention, published in 2020) [37]. The individual PRECIS‐2 scores for each study can be found in supplementary material (Table S4).

PRECIS‐2 Domains

Figure 2 shows the PRECIS‐2 wheel representing the average score for the 10 domains (9 PRECIS‐2 domains + control domain) of the 54 RCTs. The most pragmatic domains were organization (3.86 ± 1.10), recruitment (3.51 ± 1.20) and primary outcome (3.48 ± 1.33), followed by setting (2.94 ± 1.15), control (2.83 ± 1.45), flexibility—delivery (2.78 ± 0.66) and primary analysis (2.67 ± 1.90). The least pragmatic domains were follow‐up (2.52 ± 1.07), flexibility—adherence (2.45 ± 0.87) and eligibility (2.28 ± 0.79).

Inter‐rater reliability for PRECIS‐2 scoring was overall good to excellent for eight domains, with lower agreement observed for setting (ICC = 0.33) and flexibility‐delivery (ICC = −0.32) (Table S6).

Subgroup Analysis

Subgroup analyses are presented in Tables 4 and 5. The geographical location of the studies, categorized by continent, was found to significantly influence the overall pragmatism score of the clinical trials (p = 0.049). Specifically, three PRECIS‐2 domains showed significant variation depending on the ‘setting’ (p = 0.029), ‘follow‐up’ (p = 0.014) and ‘control’ (p = 0.042). In the ‘setting’ domain, studies conducted in Asia were more pragmatic, whereas those from America were more explanatory. For the ‘follow‐up’ domain, trials from Europe demonstrated higher levels of pragmatism, in contrast to those from America, which were more explanatory. The ‘control’ domain was rated as more pragmatic in studies from Asia and more explanatory in those grouped under the ‘Other’ category, which includes Oceania and global studies. Overall, studies conducted in Asia showed the highest levels of pragmatism.

The definition of sarcopenia used also significantly influenced the overall pragmatism score (p = 0.023). Two domains were notably ‘eligibility’ (p = 0.031) and ‘control’ (p = 0.005). The ‘eligibility’ domain was assessed as more pragmatic when the EWGSOP criteria were applied, while other definitions (e.g., FNIH) were associated with a more explanatory approach. Regarding the ‘control’ domain, studies using definitions other than AWGS and EWGSOP scored as more pragmatic, whereas those using EWGSOP were more explanatory. Overall, studies applying the Asian Working Group for Sarcopenia definition showed the highest levels of pragmatism.

In contrast, the type of intervention did not significantly impact the overall pragmatism score (p = 0.271). However, the ‘control’ domain again showed significant differences depending on the intervention type (p < 0.001). More specifically, the subgroup ‘other’ interventions (e.g., acupuncture) were associated with higher levels of pragmatism, while pharmacological trials were more explanatory. No other subgroup characteristics were associated with statistically significant differences, either in the overall pragmatism score or within any individual PRECIS‐2 domain.

Sensitivity Analyses

Original PRECIS‐2 Tool (Without ‘Control’ Domain)

When the control domain, which is not part of the PRECIS‐2 tool [27, 28, 29], was excluded from the analysis, the mean overall pragmatism score across the nine standard PRECIS‐2 domains was 2.95 (SD 1.28). The ranking of domains remained unchanged.

Risk of Bias Assessment

Excluding the seven studies assessed as having a high risk of bias, the average PRECIS‐2 score across all 10 domains for the selected RCTs was 2.96 (SD 1.32) (Table S7).

Discussion

This systematic review is the first to comprehensively assess the level of pragmatism in interventional sarcopenia studies and identify gaps and potential rooms of improvement to enhance the clinical relevance and feasibility of future trials in the real world. The findings reveal an overall balance between explanatory and pragmatic features across the 54 RCTs included. These trials tended to be both pragmatic and explanatory, with an average PRECIS‐2 score of 2.93 out of 5 (where 1 is highly explanatory and 5 is highly pragmatic). This suggests that, while they remain limited in their generalizability to everyday clinical practice, some areas are already pragmatic [28]. The most pragmatic domains were organization, recruitment, and primary outcome. These domains reflected actual care conditions, both in terms of the resources mobilized and the relevance of the criteria assessed. Conversely, follow‐up, flexibility‐adherence and eligibility received the lowest scores and were considered less pragmatic and more explanatory, primarily due to excessive follow‐up, rigid adherence strategies and numerous exclusion criteria.

Of the included studies, those conducted in Asia and using the AWGS definition were significantly more pragmatic. These results may be explained by the nature of the interventions in this continent. Indeed, most of the proposed interventions (i.e., 65%) were exercise‐based, and while differences by type of intervention were not statistically significant, exercise programs generally allow for greater flexibility in areas such as control and eligibility. Notably, a significant difference was observed in the ‘control’ domain (p < 0.001), which varied according to intervention type (p < 0.001). Pharmacological trials tended to be more explanatory in this domain. They often relied on placebo controls or highly standardized conditions for regulatory and safety reasons. Because of the lack of approved pharmacological treatments for sarcopenia [15], a low proportion (7.4%) of pharmacological trials was identified. Indeed, most of the molecules currently evaluated are still in Phase I or II [15]. The small number of pharmacological trials makes it difficult to evaluate the level of pragmatism of this type of intervention. In contrast, non‐pharmacological interventions, such as acupuncture or other unclassified methods, more often used comparators aligned with usual care. Importantly, adding the ‘control’ domain to the PRECIS‐2 tool accounted for this difference, as recommended in the literature. This enabled a more detailed assessment of the extent to which control conditions reflect routine clinical practice.

RCTs remain at the top of the clinical research evidence pyramid thanks to their ability to establish a link between intervention and beneficial effects on symptoms or disease [38, 39, 40]. The strict nature of RCTs remains essential to enhance both the internal validity of the study and the safety of participants, but there is room for improvement in certain domains to bring clinical trials closer to patient reality and real‐life care conditions [23, 41]. The need for pragmatism may vary depending on the type of intervention. While pharmacological trials require controlled conditions to ensure safety and regulatory compliance, non‐pharmacological interventions can more easily incorporate pragmatic elements related to adherence or eligibility. This is especially true in sarcopenia research, where patients frequently present with multimorbidity [17, 42, 43] and polypharmacy [44, 45]. In sarcopenia research, individuals who have difficulty adhering to exercise programs, whether due to physical limitations or other reasons, or those with severe forms of the disease, such as clinically significant impairments in physical performance, major mobility limitations, or very low scores on physical performance assessments, are often excluded from clinical trials. In such contexts, explanatory trials that exclude complex or vulnerable populations may produce results that fail to translate into clinical effectiveness [23]. For example, an exercise‐based intervention may appear highly effective in an RCT setting with close supervision and strict adherence protocols, but its impact may diminish substantially in real‐world conditions where support is limited and long‐term engagement is uncertain.

To address this translational gap, sarcopenia intervention studies could adopt more pragmatic approaches. For example, and depending on the type of intervention, adherence can be better reflected by favouring intention‐to‐treat analyses, eligibility can be broadened by reducing exclusion criteria when ethically feasible, and follow‐up can be lightened by limiting the number of visits. These modifications could make the trials more representative of real‐life clinical practice and more pragmatic.

There are also alternatives that may present both strengths and limits. For example, quasi‐randomized controlled trials (quasi‐RCTs) could be seen as an alternative for conducting more pragmatic research, especially for non‐pharmacological interventions. Although they do not involve the strict randomization of conventional RCTs [46] and have a lower level of internal validity, quasi‐RCTs offer other advantages. They allow for the inclusion of broader patient populations, including those often excluded from traditional RCTs [46]. Furthermore, quasi‐RCTs prioritize assessing effectiveness over efficacy, which is usually evaluated under ideal conditions [46]. Their higher external validity and ability to generalize findings to the overall population make them valuable tools for evaluating interventions in complex healthcare environments that closely reflect real‐life patient settings [47]. These characteristics make quasi‐RCTs a worthwhile option when designing clinical trials for sarcopenia.

Furthermore, RCTs would gain from not being considered in isolation. It is important to combine their results with those of observational studies [48], which often have less restrictive inclusion criteria and better reflect the diversity and complexity of real‐world patients. Observational studies are important for evaluating safety outcomes and long‐term effects, which are not fully captured in RCTs [49]. Combining data from RCTs and observational studies provides a more comprehensive understanding of an intervention's efficacy in routine clinical settings [48, 49]. This integrated approach allows for more nuanced, contextualized decision‐making in clinical practice and public health policy [49].

The growing emphasis on real‐world inclusion is reflected in the guidelines issued by regulatory authorities. Organizations such as the European Medicines Agency (EMA) and the U.S. Food and Drug Administration (FDA) now emphasize the importance of using real‐world data and designing clinical trials that better reflect medical care in clinical practice. The FDA encourages broadening inclusion criteria and adopting more inclusive recruitment practices [50]. Similarly, the EMA recognizes the importance of real‐world evidence to generate complementary data under real‐life conditions [51]. These changes indicate an evolution toward patient‐centred research. Clinical trials are now designed to not only test efficacy but also address the real needs, conditions, and diversity of patients in real life.

Limitations and Strengths

This review has several limitations that warrant consideration. Although the evaluation was conducted with methodological rigour, the critical assessment of the pragmatism domains inherently involves a degree of subjectivity. To reduce this potential bias, two independent reviewers carried out the evaluation, and discrepancies were resolved through consensus. Importantly, inter‐rater reliability was good to excellent in 8 out of the 10 domains, which indicates that both reviewers generally evaluated the domains in a consistent manner. However, a few domains showed poor inter‐rater reliability, highlighting the inherent subjectivity of the PRECIS‐2 tool, even when standardized domain definitions are applied. Additionally, the PRECIS‐2 tool does not allow for the assignment of decimal scores, which restricts evaluations, especially for tests between two levels of pragmatism. Another limitation was the lack of data in several of the included studies, which prevented the scoring of some domains and may have led to an underestimation of the true level of pragmatism. In line with Zwarenstein et al.’s recommendations [29], such domains with missing or insufficient information were deliberately left blank rather than being estimated, to preserve the integrity of the assessment. Another limitation may be related to the limited number of databases used (MEDLINE, Embase and CENTRAL), which may have restricted the comprehensiveness of the review and reduced the likelihood of identifying trials published exclusively in other databases. Similarly, excluding grey literature may have introduced publication bias by omitting studies with negative or neutral results that were not published. While this approach aligns with Cochrane’s methodological recommendations, it remains a potential limitation to consider when interpreting our results. Despite these limitations, the study presents several important strengths. This is the first systematic review to specifically assess the level of pragmatism in sarcopenia intervention studies, thereby providing an original and valuable contribution to the field. Additionally, the analysis includes the ‘control’ domain, as recommended in the literature [29], thereby enhancing the comprehensiveness of the evaluation. One of the key strengths of this review is the systematic assessment of risk of bias using the Cochrane RoB 2.0 tool. Beyond simply describing the overall risk, we conducted a sensitivity analysis by excluding studies deemed to be at high risk of bias (n = 7) to determine whether these trials had an influence on the PRECIS‐2 results. Notably, excluding these studies did not change the average overall score, but it did change the ranking slightly (recruitment and primary outcome were reversed). The ranking of the other domains has not changed. Finally, this systematic review offers practical guidance for improving future clinical trial designs by identifying commonly low pragmatic domains and proposing concrete directions for their improvement.

Conclusion

This systematic review highlights the moderate level of pragmatism in sarcopenia intervention studies, positioning these trials along the pragmatic–explanatory continuum. By identifying key methodological shortcomings, particularly in relation to eligibility, delivery adherence and follow‐up, this work provides valuable insights to improve the design and real‐world relevance of future clinical trials, and in particular for non‐pharmacological interventions. More specifically, enhancing pragmatism could be achieved by broadening inclusion criteria, simplifying follow‐up, and favouring intention‐to‐treat analyses. These findings support the development of interventions that are more patient‐centred and easier to implement, contributing to better integration of evidence into clinical practice.

We consider this systematic review as a piece of evidence that can serve as a basis for developing practical recommendations. A valuable next step would be to establish, for example, a working group including researchers, trial designers, geriatricians, physiotherapists, dietitians and older adults living with sarcopenia. Such an initiative would ensure that future sarcopenia trials are informed not only by existing literature but also by expert consensus and patient perspectives, ultimately improving their relevance and impact.

Author Contributions

C.B., D.S.R. and Z.B. contributed to the study development and design. C.B. and D.S.R. contributed to the bibliographic search, data analysis and interpretation, as well as the writing of the draft. C.B., Y.M.C. and D.S.R. contributed to study selection, while S.V.H. and Z.B. were responsible for data extraction. S.V.H. and Y.M.C. assessed the risk of bias. S.V.H., C.B. and D.S.R. contributed to the final analysis and manuscript preparation. All authors revised the article for important intellectual content and provided their final approval of the submitted manuscript. All authors read and agreed to the final version of the manuscript.

Funding

SVH is supported by a fellowship from the FSR (Fond Spécial de la Recherche) at the University of Namur.

Ethics Statement

This is a systematic review; for this study type, Ethics Committee approval is not required.

Consent

For this type of study, formal consent is not required.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information