Hospital -based Health Technology Assessment for AI-based Digital Health Technologies in Austria: Needs Exploration, Prioritization and Rapid Exploration Reports

Oliver Bernecker

Duration: Q2 2026 – Q1 2027
Language: Englisch (with German Summary)

Background: Hospital-based Health Technology Assessment processes (HB-HTA) differ from HTA conducted by national agencies in that the assessment of a health technology considers the specific care context (e.g. infrastructure, human resources, etc.) of the hospital or hospital network [1][2]. This context-specific focus aims for a pragmatic and time-efficient approach towards the evaluation of health technologies, tailored to the hospital´s individual procurement cycles [2]. Faced with increasingly complex technologies - including digital health technologies (DHTs) based on artificial intelligence (AI-DHTs), the hospital´s procurement processes also become increasingly challenging. The large number of submissions for evaluation within the AIHTA´s recent annual work programme in this technological class highlights the need for a practical and timely approach for benefit-assessment. A multitude of technology-inherent features, such as  the rapid technological evolvement, their technical life cycle, safety and security issues, increasing autonomy and complexity, complicates existing HTA and procurement processes [3].

Initial steps have been taken, internationally and by the AIHTA, to establish guidelines for assessment and evaluation of AI-based digital health technologies. However, the rapid evolution of these technologies necessitates the continuous development and refinement of procurement guidelines tailored to the specific needs and structural conditions of regional hospitals in Austria [2].

This project aims to develop tools for a structured prioritization process of AI-DHTs taking into account the specific needs of the hospital environment and building upon established hospital-based HTA (HB-HTA) processes.

Project Objectives:

Three interdependent goals will be pursued within this project:

1. Identification of existing HB-HTA structures in Austrian hospitals, needs exploration of hospital-internal procurement processes and prioritization of a selection of submitted AI-based digital health technologies within the annual work programme of the AIHTA.
2. Development of tools for rapid initial classification of AI-based digital health technologies within hospital procurement processes to support prioritization (“Topic Identification and Selection Process,” TISP).
3. Application and piloting of the developed concepts using exemplary templates, so called “Rapid Exploration Reports” (RERs), and testing their practical feasibility based on 2–3 selected submissions.

Non-objectives:

The project will not provide benefit assessments of individual AI-DHTs, nor will it conduct a systematic search for available AI-DHTs. It will also not include evaluations of AI-DHTs in outpatient or private practice settings, nor will it involve a comprehensive legal or ethical analysis of AI systems.

Research questions:

RQ1: Which needs regarding the assessment and prioritization of AI-DHTs during the procurement process exist among Austrian hospital operators and which requirements for evaluation tools can be derived from these needs?

RQ2: Which structure and content are required for the prioritization scheme and the developed tools to enable a standardized, rapid initial classification of AI-DHTs and to support decision-making within the prioritization process?

RQ3: How can existing international and AIHTA publications on HTA-Frameworks and tools for the assessment of AI-DHTs be used as a methodological basis for developing an HB-HTA specific prioritization scheme in Austria?

Methods: Methodologically, the project draws on international frameworks [4]and previous AIHTA publications [5][6][7]. To develop the RER Templates, a co-design approach involving collaboration with the points of contact within the hospital associations’ submission system (AI-Boards) is planned. The project applies both participatory and desk-research-based methods.

Needs exploration: A workshop with decision-makers from hospital associations is held to identify and reflect on existing HB-HTA structures and to present 10 submitted AI-DHTs. The workshop will include a presentation and subsequent discussion of existing AIHTA publications from 2024/2025 [5][6][7]. The documentation of the identified needs and the subsequent prioritization of relevant AI-DHTs form the basis for the next steps (RQ1). The developed content will be presented in a follow-up workshop, where the developed tools (RERs) and concepts will be reviewed and stakeholder feedback will be collected.

Development of instruments and practical testing: The content and structure of the prioritization scheme will be developed through both the adaptation and application of existing AIHTA publications (2024/2025) on international HTA frameworks and tools for the assessment of AI-DHTs, as well as the incorporation of narrative feedback from workshop participants (RQ2). This will provide the basis for the development of the RERs, which are designed to support hospitals in the initial classification of AI-DHTs (e.g. determining whether an in-depth assessment is required or whether a decision can be made without evidence analysis). Current developments in the assessment of digital health technologies (e.g. ASSESS DHT) will be integrated into the process where relevant and not already covered by previous work (RQ3).

The conceptualization of the prioritization scheme and the pilot testing of the RER templates developed based on this scheme will be conducted in close collaboration with partners from the hospital associations. This collaborative process will involve continuous dialogue with hospital representatives through meetings and group discussions, both within the aforementioned workshops and via a dedicated communication platform (email forum) established for this purpose and maintained throughout the project duration.

The RER templates will be used to pilot the developed concepts in practice. Specifically, AIHTA will develop exemplary RERs for 2–3 AI-DHTs selected from the previously presented submissions (see Workshop 1). These RERs are intended to provide hospitals with a structured evaluation pathway to support future procurement decisions regarding AI-DHTs. The final content of the RERs will be determined using the nominal group technique. Subsequently, the results will be documented and published.

References:

[1]         Di Bidino R., Lipska I., Kukla M., von Pinoci M., Papavero S. C., Marchetti M., et al. The potential of the hospital-based Health Technology Assessment: Results of a world-wide survey. Int J Technol Assess Health Care. 2025;41(1):e19. Epub 20250318. DOI: 10.1017/S0266462325000108.

[2]         Sampietro-Colom L., Lach K., A. C., Kidholm K., Pasternack I., Fure B., et al. AdHopHTA Handbook. A handbook for hospital-based Health Technology Assessment (HB-HTA). 2015.

[3]         Farah L., Davaze-Schneider J., Martin T., Nguyen P., Borget I. and Martelli N. Are current clinical studies on artificial intelligence-based medical devices comprehensive enough to support a full health technology assessment? A systematic review. 2024.

[4]         Bowie D., Needham-Taylor A., Garrett Z. and Chalkido A. Manual for piloting assessment methods for digital health technologies (interim/piloting version of the full manual D4.6). 2025.

[5]         Riegelnegg M., Giess D. and Goetz Gregor. Artificial Intelligence in Health Care with a Focus on Hospitals: Methodological Considerations for Health Technology Assessment: A Scoping Review. 2024.

[6]         Erdos J. and Grabenhofer L. Artificial Intelligence for Hospital Documentation Support. A Scoping Review of Current Use Cases. AIHTA Project Report No.: 171a; . 2026.

[7]         Erdos J. and Grabenhofer L. AI-supported Chest X-Ray Analysis for Lung Cancer Detection. Systematic Review of clinical outcomes and organisational implications. AIHTA Project Report No.: 171b;. 2026.