To map publication growth, collaboration, knowledge structure, and thematic change in artificial intelligence (AI) research in primary health care (PHC) and general practice and assess whether attention has expanded from model development to validation, implementation evaluation, and governance.

PHC AI research covers risk stratification, screening, diagnostic support, clinical decision support, and workflow optimization. However, publication growth and changing hotspots do not directly indicate clinical implementation or practical value.

Web of Science Core Collection was searched for AI- and PHC-related records published from 1 January 2015 to 19 October 2025. Of 1,481 records, 1,261 articles and review articles were screened using titles, abstracts, and author keywords, Python-developed term lexicons, weighted scoring, and manual adjudication. Records lacking a specific PHC application, focused on public or consumer health, or insufficiently relevant were excluded. The main sample included 1,123 papers for bibliometric, collaboration, thematic, citation, and bibliographic coupling analyses. A supplementary search identified 599 candidates to assess omissions due to terminological variation; 142 remained after 457 exclusions and were added to form a 1,265-paper sensitivity corpus used only for keyword-frequency, co-occurrence, and annual trend reruns. Annual output was assessed for 2015 to 2024; 2025 was retained as an incomplete-year observation but excluded from complete-year comparisons.

Annual output rose from 6 papers in 2015 to 220 in 2024 and increased continuously after 2018. Sensitivity analyses confirmed a stable thematic structure centred on AI, machine learning, PHC, electronic health records, screening and diagnosis, with overall increases in standardized AI- and PHC-term frequencies. Bibliographic coupling identified multiple application settings and model-evaluation topics. Model validation became more common; implementation/workflow remained visible, governance/regulation emerged after 2018, and fairness/bias and explainability were more prominent from 2022 to 2024. These findings indicate broader research attention, but do not demonstrate routine AI use or improved clinical outcomes.

The Amsterdam Instrumental Activities of Daily Living Questionnaire (A-IADL-Q) is well validated and commonly used to assess difficulties in everyday functioning regarding dementia. To facilitate interpretation and clinical implementation across different European countries, we aim to provide normative data and a diagnostic cutoff for dementia.

Cross-sectional data from Dutch Brain Research Registry (N = 1,064; mean (M) age = 62 ± 11 year; 69.5% female), European Medial Information Framework-Alzheimer’s Disease 90 + (N = 63; Mage = 92 ± 2 year; 52.4% female), and European Prevention of Alzheimer’s Dementia Longitudinal Cohort Study (N = 247; Mage = 63 ± 7 year; 72.1% female) were used. The generalized additive models for location, scale, and shape framework were used to obtain normative values (Z-scores). The beta distribution was applied, and combinations of age, sex, and educational attainment were modeled. The optimal cutoff for dementia was calculated using area under receiver operating curves (AUC-ROC) and Youden Index, using data from Amsterdam Dementia Cohort (N = 2,511, Mage = 64 ± 8 year, 44.4% female).

The best normative model accounted for a cubic-like decrease of IADL performance with age that was more pronounced in low compared to medium/high educational attainment. The cutoff for dementia was 1.85 standard deviation below the population mean (AUC = 0.97; 95% CI [0.97–0.98]).

We provide regression-based norms for A-IADL-Q and a diagnostic cutoff for dementia, which help improve clinical assessment of IADL performance across European countries.