Development and implementation of ICF-based functioning assessment in polish medical rehabilitation: a comprehensive system‑level pilot study

Abstract

Objective

The International Classification of Functioning, Disability and Health (ICF) provides a biopsychosocial framework for describing patients’ functioning, but its comprehensive structure poses challenges for routine clinical use and health system reporting. In Poland, rehabilitation services lacked a standardized approach to functioning assessment that would support coordinated care and comparable outcome documentation at national level. This paper aims to describe the development and implementation of a tool for the quality assessment of therapeutic rehabilitation in Poland, focusing on the standardisation of patient functioning assessment using the International Classification of Functioning, Disability, and Health (ICF).

Design

The study was structured around the creation of functioning assessment standards (categorical profiles), a unified system for planning patient support (intervention sheet), and post-rehabilitation assessment sheets to ensure comprehensive evaluation of rehabilitation effectiveness at individual, hospital, and national levels.

Methods

Within a national implementation project, we developed an ICF‑based functioning assessment system for inpatient and outpatient rehabilitation, including categorical profiles for major rehabilitation types, auxiliary tools (Short Intuitive Descriptions, rating guides, conversion protocols) and a dedicated IT application. The system was pilot‑tested in 12 rehabilitation centres (1 800 patients; 1 200 inpatients and 600 outpatients) using pre‑ and post‑rehabilitation assessments and WHO’s WHODAS 2.0 questionnaires.

Results

The pilot indicated that ICF‑based categorical profiles were feasible to use in routine practice and sensitive to changes in patients’ functioning over the course of rehabilitation. The system differentiated outcomes between centres and showed moderate to high inter‑rater agreement for most ICF categories, with several codes reaching very high ICC values, which require cautious interpretation. WHODAS 2.0 scores were broadly consistent with ICF‑based profiles, particularly in activity and participation domains. Conclusion: This pilot suggests that an ICF‑based functioning assessment system can support more standardized, biopsychosocially oriented rehabilitation in Poland, while highlighting important implementation challenges, including resource demands, staff training needs and difficulties in assessing environmental factors. Further research is needed to establish long‑term outcomes, refine psychometric properties and evaluate cost‑effectiveness of system‑wide implementation.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12913-026-14696-8.

Plain language summary

The aim of this article is to present how a tool for assessing the quality of rehabilitation was developed and implemented in Poland. The tool was based on the International Classification of Functioning, Disability and Health (ICF). The project involved the creation of functioning assessment standards (categorical profiles), a system for planning patient support (intervention sheet), and tools for evaluating the effectiveness of rehabilitation. The methodology focused on adapting selected ICF sets to Polish conditions and the needs of patients, as well as developing straightforward guides for different specialists. The implementation of the ICF system has resulted in the standardisation of patient assessments, improved communication between teams, and more precise documentation of outcomes. The project received support from key health institutions and funding from the European Social Fund. The introduction of ICF tools is intended to improve the quality and effectiveness of rehabilitation in Poland.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12913-026-14696-8.

Introduction

The International Classification of Functioning, Disability and Health (ICF), endorsed by the World Health Organization in 2001, offers a comprehensive biopsychosocial framework for describing functioning and disability across body functions and structures, activities and participation, and environmental factors. Unlike the International Classification of Diseases, which focuses on diagnosis and mortality, ICF emphasizes patients’ functioning as a third core health indicator [1–6].

In Poland, the use of ICF in rehabilitation has remained limited despite increasing international experience, partly due to the theoretical complexity of the classification and the absence of integrated system‑level solutions for documenting functioning. Existing rehabilitation reporting is largely procedure‑based and does not systematically capture functioning outcomes, which hampers coordinated care, benchmarking and strategic planning [7, 8].

At the same time a need appeared to develop user-friendly ICF tools, including simplified descriptions and rating guides for the ICF Rehabilitation Set adapted to local practice. These tools should be designed to facilitate adoption by diverse healthcare professionals, including physicians, physiotherapists, occupational therapists, psychologists, and social workers, mitigating the complexities of the original ICF framework through intuitive operational definitions and criteria [9].

The projected benefits of implementing ICF classification in Polish medical rehabilitation revolve around fostering holistic patient evaluation, improving rehabilitation qualification procedures, and supporting outcome evaluation through standardized measures of functioning [8]. These advances are expected to strengthen interdisciplinary collaboration, facilitate data comparability nationally and internationally, and ultimately improve the quality and efficiency of rehabilitation services. These efforts have been acknowledged by institutions responsible for management and financing of the national social security system, including the National Health Fund, which designed the project “Development and implementation of a comprehensive functioning assessment system in rehabilitation using foreign experiences and solutions” that has been qualified for financing from the European Social Fund (POWR.04.03.00-00-0025/19) and implemented in partnership of Katholische Universität Eichstätt-Ingolstadt in Eichstätt and the Institute of Rural Medicine in Lublin.

Study aim

The aim of this study was to develop and pilot‑test an ICF‑based functioning assessment system for Polish medical rehabilitation. functioning assessment The design includes development of functioning assessment standards (categorical profiles) assumed to enable a uniform description of patients who receive rehabilitation services both inpatient and outpatient, introduction of a unified system for planning patient support (intervention sheet), assessment of the effectiveness of planned support (post-rehabilitation assessment sheet) to ensure a comprehensive assessment of the quality and effectiveness of the support provided, both at the individual and hospital levels, as well as in all participants in the rehabilitation process.

Methodology

Study design and setting

We conducted a national implementation project combining developmental work on ICF‑based tools with a multicentre pilot study in routine rehabilitation practice. The project was funded by the European Social Fund and coordinated in partnership between the National Health Fund, the Institute of Rural Health in Lublin and an international academic partner.

Participants and rehabilitation centres

Rehabilitation centres were selected through a competitive procedure organized by the Institute of Rural Health, aiming to ensure representation of different rehabilitation types and geographical regions. The final sample included 12 centres (exceeding the initially planned 10), which collectively recruited 1 800 patients (1 200 inpatients and 600 outpatients) undergoing general, neurological and cardiopulmonary rehabilitation, as well as ambulatory services.

The multidisciplinary staff involved in the pilot consisted of physiotherapists (44%), physicians (29%) and nurses (27%), with an average professional experience of 16.3 years.

Development of ICF‑based categorical profiles

The project’s main objective was to develop a functioning assessment system applicable in Polish rehabilitation centres, both inpatient and outpatient, utilizing the ICF classification. Because ICF includes over 1400 categories covering all health-related domains, its application in an unaltered form is not feasible, which is the basic rationale behind this project.

The development of ICF core sets adapted to the context of Polish medical rehabilitation services scheme followed a systematic four-stage process:

1. Preparatory Stage: Literature review and selection of basic ICF category sets (ICF Generic-30 Set) in accordance with international standards.

The preparatory phase included systematic literature reviews of qualitative research, healthcare specialists interviews, and empirical clinical studies. Based on these, the ICF Generic Set (Minimal Generic Set, ICF Generic-7) was selected as the baseline standard [10]. This set consists of seven codes: three related to body functions and four related to activities and participation. Despite the compact size, it sufficiently represents significant functional differences across diverse health conditions with strong psychometric properties. The ICF Rehabilitation Set (ICF Generic-30 Set) extends the Generic-7 by adding 23 categories: six addressing body functions and seventeen related to activities and participation. This expanded set was recommended as the standard for assessing and reporting functioning and disability in healthcare systems [11]. Additional condition-specific core sets were developed for acute and post-acute phases in neurological, cardiovascular, musculoskeletal, and geriatric rehabilitation, with the ICF Generic-7 serving as the primary assessment basis for acute rehabilitation settings [12]. An international expert panel used consensus voting to select essential categories; those accepted by 75% or more were definitively included, those with the level of acceptance between 40% and 75% were re-discussed, while the rest was excluded.

• 2.Main Development Stage: Multidisciplinary expert panels, profile development for specific rehabilitation fields, and category selection/modification.

A multidisciplinary expert team selected ICF categories enabling functioning assessment of different health conditions, based on the data reflecting rehabilitation service usage obtained from Poland’s National Health Fund for years 2018–2020. Categorical Profiles were developed for main rehabilitation types:

• General inpatient systemic rehabilitation: foundation profile incorporating ICF categories integral to functional diagnosis of body systems;

• Neurological inpatient rehabilitation: combined general profile with targeted codes capturing cognitive, communicative, and behavioral aspects specific to neurological conditions;

• Cardiopulmonary inpatient rehabilitation: Unified profile acknowledging clinical overlap between cardiovascular and respiratory diseases, integrating functional domains and categories such as muscle endurance and dyspnea;

• General outpatient profile: Deliberately includes ICF codes relevant to community mobility, employment, recreation, and independent living, reflecting broader patient activities and environmental challenges outside hospital settings.

While international condition‑specific ICF Core Sets were available, additional disease‑specific categorical profiles were developed to reflect the case mix of Polish rehabilitation services and financing rules, and to ensure sufficient clinical granularity for routine decision‑making. These profiles were explicitly linked to existing Core Sets, with clear documentation of which categories originated from previous international work and which were proposed within this project (see Annex 1, Supplementary Tables 1–4).

All profiles incorporate a unified environmental factor set based on the “Minimal Set of Environmental Factors”, not to omit the crucial role of environmental context in enabling and constraining social and daily life participation. The full set is recommended for use in all functioning assessments and is presented in details in the Annex 1 (Supplementary Table 5).

The proposed profiles have been subject to consultation with the international partners, who provided expertise with important insights into the developed proposal. The expertise confirmed that ICF categorical profiles designed in relation to disease groups and healthcare contexts was a reasonable approach. However, there was a number of key considerations outlined. First, the expertise suggested future data analyses to address ICF category sensitivity for measuring progress and differentiating between patients with varying functioning levels. Secondly, partners stressed that the new system should systematically demonstrate differences from current systems in terms of therapeutic orientation, otherwise, it will merely represent a translation of the old system into the ICF nomenclature. Thirdly, certain ICF categories have been highlighted as potentially appearing as redundant and though being subject to removal. It should be noted however, that the degree of change cannot be the only criterion for removing a category, since even categories that cannot be modified may contain information of great clinical importance. Fourth, implementation of the ICF in outpatient sector has been marked as especially challenging, due to patient heterogeneity, extensive documentation requirements, and the requirement for significant therapeutic involvement.

An important aspect of the entire process, as well as the selection of technical tools used in the process of verifying the correctness of categorical profiles, was to rely on tools available and validated in the Polish language, due to its phonetic and articulatory specificity. A good illustration of this need is the categorical profile covering articulation functions that are important in neurological rehabilitation. Diagnosing pronunciation norms must be related to the assessment of the realisation of all vowel and consonant phonemes of the Polish language. The manner of realisation of all phonemes of the language should be consistent with the accepted phonological inventory of both basic and positional variants. The examination of articulation functions should cover the articulation of all vowels and consonants in various contexts: neutral – in order to describe basic variants and identify them as normative or impaired, as well as in consonant groups in order to describe the process of coarticulation in various combinations (in isolation, in syllables, in logatomes, in words with varying numbers of syllables, in sentences, in texts) and positions (interword and intraword, including initial, medial and final positions).

Assessment methodology and implementation

ICF qualifiers are essential for assessing severity and changes in patient functioning. Recognizing limited interrater reliability challenges, auxiliary tools were developed and adapted to the Polish context. First of these were Short Intuitive Descriptions (SIDs) – namely a simplified category descriptions facilitating practical use of ICF. The second were Rating Reference Guides constituting a standardized criteria supporting consistent qualification. Finally, Conversion Protocols were developed to guide conversion from established clinical tools (e.g., Berg Balance Scale, Visual Analog Scale) to appropriate ICF qualifiers.

Clinical Assessment Methods included Observational Scales in a form of four-point scales ranging from “performs independently and safely” to “unable to perform” for activities and participation codes; Self-Report Tools to capture patient/interviewer-reported data using the same conceptual framework as clinical observation; Standardized Clinical Tests in a form of FOUR Score (consciousness), Clock Drawing Test (executive function), Berg Balance Scale (vestibular function), NRS (pain), MRC scale (muscle strength), and Ashworth Scale (muscle tone).

Assessment use is advised to be flexible with adaptation to patient context. Clinicians have been encouraged to supplement profile-based assessment with additional relevant codes and domains, particularly for complex or comorbid cases.

• 3.Testing Stage: Tool preparation, staff training, pilot implementation, supervision, feedback collection, and recommendations development.

Following development of assessment procedures, practical testing was conducted through pilot studies involving comprehensive functioning assessments. The Institute of Rural Health (IMW Lublin) announced a competition to select rehabilitation institutions for pilot implementation focusing on different rehabilitation fields. Eventually, the piloting included 12 selected centers (exceeding the planned 10), 1,800 total patients (1,200 inpatients and 600 outpatients), professional composition of 44% physiotherapists, 29% physicians and 27% nurses, with average professional experience of 16.3 years in all groups.

The implementation process of the pilot study included:

1. Pre-pilot stage consisting of mandatory staff training on ICF principles, categorical profiles, and IT tool usage;

2. Main stage of dual functioning assessments (pre- and post-rehabilitation) with dedicated software recording, ongoing supervision, and case discussions;

3. Post-pilot stage, namely a standardized collaborative reports evaluating assessment areas and improvement suggestions.

A multidisciplinary team established evaluation protocols, linking specific ICF codes (e.g., consciousness, orientation, energy, sleep, memory, attention, emotions, pain, mobility, self-care, communication) with concrete clinical or questionnaire-based assessment tools. For each code, defined scoring systems, conversion guidelines for ICF qualifiers (0–4: no problem to severe problem), and test instructions were provided.

A FSR application was developed as a comprehensive IT system to support and standardize the assessment process. The application included modules for:

• User and patient management.

• Diagnosis entry (ICD-10 codes).

• ICF-based profile selection tailored to rehabilitation type.

• Pre- and post-intervention effectiveness assessment.

• Intervention planning and documentation.

• Evaluation and reporting (comparative analysis of outcomes).

• Self-assessment using WHODAS 2.0-based questionnaires [13].

• 4.Final Stage: Specification refinement based on pilot project findings.

Data collection

In each centre, trained multidisciplinary teams conducted ICF‑based functioning assessments using the relevant categorical profiles and auxiliary tools at admission and discharge. For a subset of cases, parallel ratings were obtained from two raters to allow estimation of inter‑rater agreement. Patients also completed WHODAS 2.0 questionnaires to provide standardized self‑reported information on functioning.

All assessments were documented using a dedicated IT application (FSR), which integrated ICF‑based profiles, pre‑ and post‑rehabilitation assessments, intervention planning and reporting modules.

Statistical analysis

We performed descriptive analyses of patient characteristics and functioning scores at admission and discharge, stratified by rehabilitation type. Inter‑rater agreement for ICF qualifiers was assessed using intraclass correlation coefficients (ICCs) for individual codes and aggregated domain scores, accompanied by 95% confidence intervals. We also examined correlations between ICF‑based profiles and WHODAS 2.0 domains to explore convergent validity. Psychometric analyses of the categorical profiles and auxiliary scales were conducted as part of the project’s broader psychometric work package and are summarized in the final report.

Results

Study population and feasibility

A total of 1,800 patients were enrolled across 12 participating centres, including 1,200 inpatients and 600 outpatients representing different rehabilitation profiles. The assessment was performed by a multidisciplinary group of 111 medical specialists, consisting of physiotherapists (44%), physicians (29%), and nurses (27%). The staff was highly experienced, with an average of 16.3 years of professional practice (nurses: 22.5 years; physicians: 15.2 years; physiotherapists: 11.2 years). Implementation feasibility was evaluated using standardized measures: the Acceptability of Intervention Measure (AIM) reached a mean score of 3.26 (SD = 1.02), the Intervention Appropriateness Measure (IAM) scored 3.13 (SD = 0.85), and the Feasibility of Intervention Measure (FIM) scored 3.14 (SD = 0.82) on a 5-point scale.

Changes in functioning and differentiation between centres

ICF‑based categorical profiles effectively detected improvements in patient functioning between admission and discharge. Statistical analysis confirmed significant functional changes in several centers, such as AHOP (p = 0.005), Międzyrzec Podlaski (p = 0.013), Nałęczów (p = 0.030), and Zakopane (p = 0.047). Improvements were most pronounced in the activity and participation domains (d-codes), particularly in centers like AHOP (p = 0.008 for this domain). The magnitude of detected change varied between facilities, reflecting differences in patient populations and rehabilitation intensity.

Inter‑rater agreement and psychometric indicators

Inter‑rater reliability was exceptionally high, with the Intraclass Correlation Coefficient (ICC) reaching 1.000 for most codes. This absolute consistency resulted from the application of rigid conversion protocols within the FSR IT system, which automatically mapped objective clinical test results (e.g., 6MWT, Barthel Index, Berg Scale) to specific ICF qualifiers, thereby eliminating subjective rating variance. Psychometric validation supported the internal consistency of the profiles, although it highlighted that environmental codes (“e” codes) were more difficult to standardize.

Patient‑reported outcomes (WHODAS 2.0)

The WHODAS 2.0 tool confirmed significant functional improvements, showing high consistency with clinician-rated categorical profiles. Statistically significant changes in patient self-assessment were recorded in centers such as Nałęczów (p < 0.001), PUM (p < 0.001), and Iwonicz (p = 0.007). Discrepancies between clinician and patient ratings were noted primarily in complex psychosocial domains, reinforcing the need for integrated assessment.

IT application and practical challenges

The FSR IT application standardized the recording of assessments for all 1,800 patients. While the system enabled transparent and auditable functional profiling, user feedback varied by profession. Physiotherapists gave the highest acceptability ratings (AIM: M = 3.51), whereas nurses provided lower scores (AIM: M = 2.83), often citing the substantial time requirements for data entry in inpatient workflows. Implementation was most successful in centers with existing digital infrastructure, such as AHOP (AIM: 4.00) and Nałęczów (AIM: 4.16).

Pilot feedback

The pilot confirmed that ICF codes grounded in robust clinical tests achieved the highest user acceptance. Approximately 12% of the testing staff held advanced scientific degrees and 26% had completed formal ICF training, which positively influenced the quality of data collection. Despite the moderate average scores for “fit into daily practice” (3.16/5), the implementation was seen as a vital step toward standardizing rehabilitation quality across Poland. Challenges remained regarding the complexity of environmental factors, which were marked for further optimization.

Based on the pilot study the implementation of ICF classification could be believed to induce a number of system-level outcomes, including strengthening interdisciplinary character of rehabilitation teams with more collaborative decision-making and, what’s even more important, improved standardization between centers and treatment types. The system may also facilitate system-wide rehabilitation planning and long-term data use. At the same time it faces a number of potential obstacles, like substantial level of resource and time requirements for comprehensive assessment, temporary reduction in staff availability for direct therapy, a need for additional funding exceeding standard clinical budgets, as well as investment in equipment and staffing.

Discussion

This pilot study suggests that an ICF‑based functioning assessment system can be integrated into routine rehabilitation practice in Poland and may support more standardized, biopsychosocially oriented care. Our findings are broadly consistent with international experiences demonstrating that ICF‑based tools can capture clinically meaningful changes in functioning and support multidisciplinary collaboration, while at the same time posing challenges related to complexity, training and documentation burden. The observed improvements in functioning scores and the differentiation of outcomes between centres indicate that the developed categorical profiles and auxiliary tools are sensitive to change and capable of reflecting differences in service delivery contexts. However, these findings stem from a pilot conducted in selected centres with additional support and funding, and therefore cannot be directly generalized to the entire rehabilitation system.The obtained results confirm the developed ICF categorical profiles to be effective in detecting changes in patients’ functional status and differentiating outcomes between centers, which is in line with previous ICF Core Sets validation studies [14, 15]. Particularly important is the confirmation of high inter-rater reliability and correlation with recognized functioning assessment tools, such as the Barthel Index. These findings are broadly comparable to those reported by Starrost et al. [16] and Kohler et al. [17] in studies exploring ICF as a potential rehabilitation outcome measure (see also [18, 19]), and fall within the range of ICC values observed in previous ICF-based research. However, given the pilot nature of our study, these comparisons should be interpreted cautiously.

The positive results regarding the WHODAS 2.0 tool, which showed consistency with ICF-based categorical profiles, confirm WHO recommendations for the use of standardized functional instruments [20]. Tempest and McIntyre demonstrated similar effectiveness of the ICF-based post-stroke rehabilitation program (ICF-PSRP) in improving functioning at the activity and participation level [21], so did Wong and colleagues [22]. Taken together with previous work, our results support the applicability of ICF-based approaches across different rehabilitation contexts, although further research is needed to confirm their performance in diverse settings.

The importance of a team-based, multidisciplinary approach to functioning assessment remains in line with existing international recommendations for rehabilitation care [23, 24], with ICF classification being potentially effective instrument of facilitating this kind of cooperation [25]. However, the results outline some challenges related to the time-consuming nature of the assessment process and the need for intensive staff training. These observations are consistent with results from studies conducted in other countries, where similar implementation barriers were identified [26, 27].

Tempest et al. in their action research study emphasized the importance of adapting the language and format of ICF to local needs and the necessity of external facilitation of the implementation process [28, 29]. Similarly, Paltamaa and colleagues showed that effective ICF implementation requires structural team training and a practical approach to multidisciplinary work [30]. Our studies confirm these observations, indicating the need for continuous educational support and clinical supervision. The effectiveness of training strategies is varying from one case to another, with educational outreach visits and workplace-based training showing superior results compared to traditional didactic approaches [24]. This also aligns with our findings that hands-on, contextual training was more effective than theoretical ICF education alone.

The financial and organizational barriers identified in the project reflect broader challenges of healthcare systems in post-communist countries. The comprehensive assessment was only possible due to additional funding beyond standard clinical budgets, which is consistent with international studies on implementation barriers in healthcare quality improvement programs [5, 31, 32]. The need for sustainable financing mechanisms represents a critical challenge that must be addressed at policy level.

These findings remain consistent with the WHO Rehabilitation 2030 initiative, which underlines the need for system-level integration of rehabilitation services and appropriate financing mechanisms to support function-centered care pathways [5, 33]. The Polish experience demonstrates that ICF implementation requires not only additional funding but also systemic organizational changes at multiple levels.

The developed categorical profiles showed varying utility depending on the type of rehabilitation, with the best results obtained in inpatient rehabilitation. These results are consistent with international experiences indicating greater implementation challenges in outpatient care settings [27]. The unified cardiopulmonary profile reflects practical clinical needs and demonstrates how ICF Core Sets can be adapted to local practice patterns while maintaining scientific rigor.

Van Leeuwen and colleagues in their analysis of ICF implementation barriers emphasized the importance of organizational context and availability of diagnostic resources [27]. Our findings support this, particularly regarding the challenges faced in ambulatory settings where resources and specialist staff availability are more limited.

The ICF-based profiles addressed biopsychosocial model needs, particularly in inpatient settings. This is consistent with recent research on biopsychosocial rehabilitation which emphasizes the importance of integrated assessment approaches that consider biological, psychological, and social factors simultaneously [34, 35]. The ability to capture environmental factors, though challenging in institutional settings, represents an important advancement toward holistic rehabilitation care.

The integration of environmental factors (e-codes) remains challenging though important, as these factors are recognized as crucial determinants of social and daily life participation and functioning outcomes [36–38]. The difficulties encountered in measuring environmental factors during institutional rehabilitation are no different from international experiences and reveal the need for continued development of measurement approaches in this domain [39].

Our study confirmed moderate to high inter-rater reliability for most ICF categories, which is consistent with international validation studies [16, 17, 40]. However, challenges with environmental codes and certain body function codes requiring significant clinical judgment show ongoing reliability concerns that have been noted in the broader ICF literature [41–43]. The development of conversion protocols from established clinical tests to ICF qualifiers represents a practical advancement that could facilitate broader ICF adoption. Also Starrost and colleagues identified similar reliability challenges and emphasized the importance of operationalizing ICF categories and standardizing the rating process [16]. Our development of Short Intuitive Descriptions (SIDs) and rating reference guides addresses these concerns and provides a model for other implementation efforts.

The developed IT system (FSR application) appears to be a significant contribution to practical ICF use, though limitations in outpatient settings indicate areas for continued technological development. Digital health solutions are increasingly recognized as essential for ICF implementation, supporting standardized assessment and enabling quality monitoring in health care settings of multiple type [8, 14, 44, 45].

The integration of WHODAS 2.0 as both a validation tool and practical assessment option demonstrates how established patient-reported outcome measures can complement ICF-based assessment systems, providing both scientific rigor and clinical utility.

This study has several limitations. First, the pilot was conducted in a limited number of centres that volunteered or were selected through a competitive process, which may introduce selection bias and limit generalizability. Second, the observation period was relatively short and focused on admission‑to‑discharge changes, without long‑term follow‑up of functioning or service use. Third, intensive training and external support were provided as part of the project, which may not be easily replicable in routine conditions and could have influenced both assessment quality and staff attitudes. Fourth, while we reported ICCs and convergent validity indicators, more comprehensive psychometric analyses, including responsiveness and cross‑setting comparability, are needed before using the system for benchmarking or funding decisions. Finally, outpatient settings were under‑represented in terms of complete data, and environmental factor codes remained difficult to rate consistently. Further research is needed on optimizing assessment tools, particularly regarding codes requiring significant clinical judgment, and developing effective training strategies for healthcare personnel. The integration of artificial intelligence and machine learning approaches may offer future opportunities to enhance ICF-based assessment and decision-making processes [46].

Conclusions and recommendations

The national pilot of an ICF‑based functioning assessment system demonstrated that standardized ICF‑based profiles, auxiliary tools and an integrated IT application can be implemented in Polish medical rehabilitation and used to monitor changes in functioning across different rehabilitation types. At the same time, the project highlighted substantial resource, training and organizational requirements, particularly in outpatient care and in relation to environmental factor assessment.

Future work should focus on refining categorical profiles and rating procedures, strengthening psychometric evidence, integrating functioning data into routine reporting and financing mechanisms, and evaluating long‑term outcomes and cost‑effectiveness of system‑wide implementation. Among the implementation recommendations the study suggest to:

1. Emphasize ICF codes grounded in robust clinical tests, while limiting environmental codes where measurement is challenging.

2. Incorporate WHODAS 2.0 for activity and participation assessment.

3. Focus initial deployment on inpatient rehabilitation settings, while addressing ambulatory rehabilitation’s IT and staffing barriers.

4. Develop training strategies for healthcare personnel.

We also recommend gradual implementation allowing outcome comparisons and integration of assessment methodology into routine clinical practice. The process should be preceded by development of conversion protocols from clinical tests to ICF qualifiers and standardized tools to support national and international benchmarking.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

ICC

Intraclass correlation

ICF

International Classification of Functioning, Disability and Health

ICF

PSRP-ICF-based post-stroke rehabilitation program

SIDs

Short Intuitive Descriptions

WHODAS 2.0

World Health Organization Disability Assessment Schedule 2.0

Author contributions

K.P. and A.F. developed a study concept and design. K.P. was the leading researcher in this study. P.R. wrote the manuscript draft. All authors reviewed the manuscript. K.K. and D.K. prepared supplementary materials.

Funding

The study was based on a results of a project: Development and implementation of a comprehensive functioning assessment system in rehabilitation using foreign experiences and solutions which was financed by the the European Social Fund (POWR.04.03.00-00-0025/19).

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Ethics committee or institutional review board (IRB) approval was not required for this study. The project had a system-level, non-interventional implementation and methodological development character and did not constitute clinical research or a medical experiment involving patients. During the pilot implementation, healthcare professionals in participating rehabilitation centers entered patient functioning assessment data into an IT system developed and administered by the National Health Fund (NFZ). These activities were conducted as part of routine clinical documentation and service provision. The authors of this study did not have access to individual-level or identifiable patient data at any stage of the project. All analyses were performed exclusively on fully anonymized and aggregated data, representing secondary use of data generated by the system. Patient consent, in accordance with applicable national regulations [47, 48] and standard clinical practice, concerned hospitalization and related medical procedures — and was unrelated to this study. It was administered locally by healthcare professionals. Participation of healthcare professionals in qualitative interviews conducted as part of the implementation process was voluntary, and informed consent was obtained from all interviewees. The study was conducted in accordance with the principles of the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.