Poisoning and envenomation linkage to evaluate outcomes and clinical variation in Australia (PAVLOVA): a longitudinal data-linkage cohort of acute poisonings, envenomations, and adverse drug reactions in New South Wales, Australia, 2011-2020.

Introduction: Poisoning is a leading cause of morbidity and mortality that is increasing in many countries. Better data are needed to understand epidemiology and outcomes of poisoning. This work describes a new poisoning data linkage cohort in New South Wales, Australia (population approximately 8 million).

Methods: This is a longitudinal health record linkage, 2011-2020, including data from: ambulance call-outs, emergency department presentations, hospital admissions, death registrations, the poisons centre, and four tertiary toxicology units. Individuals with poisoning, venomous animal/plant exposures, or adverse drug reaction events were included.

Results: There were 845,217 linkable events relating to 400,642 ambulance, 688,484 emergency department, 682,013 admission, 40,456 toxicology, and 11,879 death records. There were 572,841 people with events; the median age at the time of first event was 57 years, and 51.9% were female. Events leading to patient admission were most commonly adverse drug reactions (n = 511,263), intentional poisonings (n = 68,646), unintentional poisonings (n = 54,840) and animal/plant exposures (n = 11,092). Demographics varied by cause: intentional poisoning (median age 33 years, 61.7% female); unintentional poisoning/animals/plants (median age 43 years, 45% female); and adverse drug reactions (median age 70 years, 54% female). Adolescent females had highest rates of intentional poisoning, while unintentional poisoning had a bimodal distribution, highest in children <5 years old and males aged 20 to 50 years. Substance use disorders were documented comorbidities for 44% of intentional poisoning, 29% of unintentional poisoning, and 13% of adverse drug reaction-related admissions; mood disorders were documented for 54%, 17% and 10% of these admissions, respectively.

Discussion: Poisonings and hospitalised adverse drug reactions are common in New South Wales, affecting approximately 8% of the population in 10 years. This linkage improves understanding of poisoning risks and outcomes in Australia.

Conclusions: This novel data linkage provides a unique opportunity to study poisoning across multiple settings for an individual over an extended period.