Integrating primary and meta-analytic effect size estimates on the effects of PFAS and organochlorine pesticides on human health outcomes

Kyle Morrison , Lorenzo Ricolfi, Iwo Gross , Ayumi Mizuno, Pietro Pollo, Marija Purgar, Malgorzata Lagisz, Shinichi Nakagawa

Persistent organic pollutants such as per- and polyfluoroalkyl substances (PFAS) and
organochlorine pesticides are among the most extensively studied environmental
contaminants, yet the evidence base remains fragmented across primary studies and
meta-analyses, limiting its ability to chart new research directions. To address this
fragmentation, we developed a unified database that links primary epidemiological
estimates with meta-analytic summaries of associations between PFAS and
organochlorine pesticide exposures and human health outcomes. The database was
constructed following a systematic five-step process encompassing (1) literature
search, (2) literature screening, (3) data extraction, (4) data curation, and (5)
visualisation. The final database includes 627 unique primary studies (n) providing
3,627 individual primary study estimates (k) that were synthesised in 91 meta-analyses
comprising 457 pooled effect size estimates. The most frequently investigated
chemicals were PFOA (CAS: 335-67-1, n = 268, k = 736) and PFOS (CAS: 1763-23-1, n =
276, k = 682) among PFAS, and DDE (CAS: 72-55-9, n = 95 k = 150) and p,p’-DDE (CAS:
72-55-9, n = 90, k = 136) among organochlorine pesticides. Metabolic disorders (ICD11:
5C8Z, n = 39, k = 333) and neoplasms of breast (ICD11: 2C6Z, n = 78, k = 309) were the
most assessed health outcomes. Primary studies were most often led by authors based
in the United States, whereas meta-analyses were predominantly conducted by
researchers in China. Based on the structure of this database, we outline practical
methodological recommendations to improve future conduct and reporting practices,
and we discuss potential applications of this database to support future research in
environmental health.