Research on Artificial Intelligence and Primary Care: A Scoping Review
Bibliographic record
Abstract
ABSTRACT Objective The purpose of this study was to assess the nature and extent of the body of research on artificial intelligence (AI) and primary care. Methods We performed a scoping review, searching 11 published and grey literature databases with subject headings and key words pertaining to the concepts of 1) AI and 2) primary care: MEDLINE, EMBASE, Cinahl, Cochrane Library, Web of Science, Scopus, IEEE Xplore, ACM Digital Library, MathSciNet, AAAI, arXiv. Screening included title and abstract and then full text stages. Final inclusion criteria: 1) research study of any design, 2) developed or used AI, 3) used primary care data and/or study conducted in a primary care setting and/or explicit mention of study applicability to primary care; exclusion criteria: 1) narrative, editorial, or textbook chapter, 2) not applicable to primary care population or settings, 3) full text inaccessible in the English Language. We extracted and summarized seven key characteristics of included studies: overall study purpose(s), author appointments, primary care functions, author intended target end user(s), target health condition(s), location of data source(s) (if any), subfield(s) of AI. Results Of 5,515 non-duplicate documents, 405 met our eligibility criteria. The body of literature is primarily focused on creating novel AI methods or modifying existing AI methods to support physician diagnostic or treatment recommendations, for chronic conditions, using data from higher income countries. Meaningfully more studies had at least one author with a technology, engineering, or math appointment than with a primary care appointment (57 (14%) compared to 217 (54%)). Predominant AI subfields were supervised machine learning and expert systems. Discussion Overall, AI research associated with primary care is at an early stage of maturity with respect to widespread implementation in practice settings. For the field to progress, more interdisciplinary research teams with end-user engagement and evaluation studies are needed. SUMMARY BOXES Section 1: What is already known on this topic Advancements in technology and the availability of health data have increased opportunities for artificial intelligence to be used for primary care purposes. No comprehensive review of research on artificial intelligence associated with primary care has been performed. Section 2: What this study adds The body of research on artificial intelligence and primary care is driven by authors without appointments in primary care departments and is focused on developing artificial intelligence methods to support diagnostic and treatment decisions. There is a need for more interdisciplinary research teams and evaluation of artificial intelligence projects in ‘real world’ practice settings.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.066 | 0.199 |
| Meta-epidemiology (narrow) | 0.002 | 0.002 |
| Meta-epidemiology (broad) | 0.007 | 0.005 |
| Bibliometrics | 0.037 | 0.043 |
| Science and technology studies | 0.003 | 0.004 |
| Scholarly communication | 0.012 | 0.010 |
| Open science | 0.003 | 0.005 |
| Research integrity | 0.006 | 0.004 |
| Insufficient payload (model declined to judge) | 0.007 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".