JMIR Publications

ABSTRACT

Background:

Concerns have been expressed about the abundance of new clinical prediction models (CPMs) proposed in the literature. The extent of this proliferation in prediction research, however, remains unclear.

Objective:

To estimate the total and annual number of CPM development publications available for all medical fields.

Methods:

Using a validated search strategy, we conducted a systematic search for prediction model studies published in Pubmed and Embase between 1995 and the end of 2020. By taking random samples for each year, we identified eligible studies that developed a multivariable model (diagnostic or prognostic) for individual-level prediction of a health outcome across all medical fields. Exclusion criteria included the development of models with a single predictor, studies not concerning humans, methodological studies, conference abstracts, articles with unavailable full text, and studies not available in English. Next, we estimated the total and annual number of published regression-based multivariable CPM development articles, based on the total number of publications, proportion of included articles and search sensitivity. Furthermore, we used an adjusted Poisson regression to extrapolate our results to years 1950-2024. Additionally, we estimated the number of articles developing non-regression-based CPMs.

Results:

From a random sample of 10,660 articles published between 1995 and 2020, 109 regression-based CPM development articles were included. We estimated that 82,772 (95%CI: 65,313-100,231) CPM development articles using regression were published, with an acceleration in model development from 2010 onward. With the addition of articles that used other techniques than regression (e.g., machine learning), the number rose to 147,714 (125,201-170,226). After extrapolation to 1950-2024, numbers increased to 156,673 and 248,431 for regression-based models and all models, respectively.

Conclusions:

Based on a representative sample of publications from the literature, we estimate that close to 250,000 articles reporting the development of clinical prediction models for all medical fields have been published until 2024. CPM development publication is ever increasing. To prevent research waste and to close the gap between research and clinical practice, focus should shift away from developing new CPMs to facilitating model validation and impact assessment of the plethora of available CPMs. Limitations of this study include restriction of search to articles available in English and development of the validated search strategy prior to the popularity of artificial intelligence and machine learning models.