JMIR Publications

ABSTRACT

Background:

Administrative health databases (AHDs) are increasingly used for population-based health research and surveillance, in particular for vulnerable such as farmers.

Objective:

To identify how AHDs are used for addressing health issues in farming populations.

Methods:

A scoping review was conducted using PubMed and Web of Science databases. To develop and validate the search strategy, previous reviews that examined the use of AHDs for population-based research were identified and refined. A total of 72 terms pertaining to farming population and AHDs were used. Twenty-nine metadata were extracted from each record included in the scoping review. To analyze the research trends on the use of AHDs for public health and epidemiological research in farming populations, a bibliometric approach was conducted. Trends and hotspots were investigated with keyword frequency, co-occurrence, and thematic map. A focus of the current research regarding agricultural/farming exposome was also performed.

Results:

From 1975-2024, 296 publications were published in 118 journals and written by 1125 authors from 34 countries, with an annual growth rate of 5.2%. The most used AHDs were electronic health/medical records (n=124, 42.3%), followed by cancer registries (n=120, 41%), insurance claims (n=106, 36.2%), population registries (n=95, 32.4%), mortality registries (n=75, 25.6%), and hospital discharge (n=41, 14.0%). There was no consensus on the choice of exposure proxies (e.g., job title, farming activity, pesticide compounds) or assessment methods (e.g., self-reported questionnaires or interviews) to assess farming exposure. Cancer, mortality, and pesticide exposure were the most studied topics. Climate change, air pollution, auto-immune diseases, and sleep disorders received less attention. AGRICAN (Agriculture and Cancer) (n=18, 6.1%) and AHS (Agricultural Health Study) (n=17, 5.8%) were the most prolific cohorts.

Conclusions:

This work allowed us to identify several research trends, gaps, and future research directions. Data from low- and middle-income countries and for some subgroups of farming populations are needed (e.g., females, migrant workers). A unique opportunity lies in the integration of person-generated data (e.g., mobile health (mHealth), social health (e.g., social media, digital footprints), and wearable sensors), as well as the use of artificial intelligence, which, to date, have been rarely performed. Because the well-recognized adverse consequences of farming exposure on human health are of high societal and scientific interest, the reuse of AHDs will most likely become even more relevant, paving the way for a new era of digital public health and epidemiology.