Public healthcare costs associated with long-term exposure to mixtures of persistent organic pollutants in two areas of Southern Spain: A longitudinal analysis

Abstract

Background: Polychlorinated biphenyls and organochlorine pesticides are persistent organic pollutants (POPs) that had been banned or restricted in many countries, including Spain. However, their ubiquity still poses environmental and human health threats.

Objective: To longitudinally explore public healthcare costs associated with long-term exposure to a mixture of 8 POPs in a cohort of residents of two areas of Granada Province, Southern Spain.

Methods: Longitudinal study in a subsample (n = 385) of GraMo adult cohort. Exposure assessment was performed by analyzing adipose tissue POP concentrations at recruitment. Average primary care (APC) and average hospital care (AHC) expenditures of each participant over 14 years were estimated using the data from their medical records. Data analyses were performed by robust MM regression, weighted quantile sum regression (WQS) and G-computation analysis.

Results: In the adjusted robust MM models for APC, most POPs showed positive beta coefficients, being Hexachlorobenzene (HCB) significantly associated (beta 1.87; 95% Confidence interval (95%CI): 0.17, 3.57). The magnitude of this association increased (beta: 3.72; 95%CI: 0.80, 6.64) when the analyses were restricted to semirural residents, where p-HCH was also marginally-significantly associated to APC (beta: 3.40; 95%CI: -0.10, 6.90). WQS revealed a positive but non-significant mixture association with APC (beta: 0.14; 95%CI: -0.06, 0.34), mainly accounted for by p-HCH (54%) and HCB (43%), that was borderline-significant in the semi-rural residents (beta: 0.23; 95%CI: -0.01, 0.48). No significant results were observed in G-Computation analyses.

Conclusion: Long-term exposure to POP mixtures might represent a modifiable factor increasing healthcare costs, thus affecting the efficiency of the healthcare systems. However, and owing the complexity of the potential causal pathways and the limitations of the present study, further research is warranted to fully elucidate ascertain whether interventions to reduce human exposure should be considered in healthcare policies.