Medicine use preceding fatal poisonings among individuals with bipolar disorder

Introduction: Poisoning is a common cause of death in patients with bipolar disorder. While pharmacotherapy may reduce overall suicide risk, it also poses a unique concern, as prescribed medicines can be causative agents in both suicidal and accidental poisonings.

The aim of this study was to investigate the association between prescription medicine use and fatal poisonings due to medicines or illicit substances among individuals with bipolar disorder.

Methods: We conducted a nationwide, nested case-control study using data from inpatient and outpatient specialized healthcare registers, medication dispensing records, causes of death register, and population registers in Finland in 1996–2018. The base cohort of this study included 60,045 individuals diagnosed with bipolar disorder (ICD-10 codes F30-F31, excluding individuals with schizophrenia-spectrum disorders. The cases were defined as the individuals who died of a poisoning due to medicines or illicit substances: ICD-10 codes F11–F19 (excluding F17), X40–X44, X60–X64, Y10–Y14 as main causes of death. The cases were matched with 4 controls by sex and age (±1 year) on the date of death of the case (index date). Controls were required to be alive on the index date and not diagnosed with schizophrenia-spectrum disorder. Prescription medicine use data was extracted from the national Prescription Register and modelled with the previously validated PRE2DUP method. The studied medicines included mood stabilizers, antipsychotics, antidepressants, benzodiazepines, benzodiazepine-related drugs (i.e., the so-called Z-drugs), gabapentinoids, opioids, paracetamol, and beta blockers. Exposure was analyzed during a two-week period preceding the index date. The association between fatal poisoning and medicine use in the prior two weeks was examined using conditional logistic regression, adjusting for sociodemographic characters, comorbidities, and other medicine use.

Results: The 777 (1.3%) cases who died of poisoning due to medicines or illicit substances during the study period were matched to 3,108 controls. Of cases, 56.5% (n = 439) died of a suicidal poisoning and 38.4% (n = 298) of an accidental poisoning. Tricyclic antidepressants (aOR 3.05, 95% CI 2.09 – 4.46), gabapentinoids (aOR 2.50, 1.73 – 3.63), opioids (aOR 2.31, 1.65 – 3.23), benzodiazepines (aOR 2.15, 1.72 – 2.67), Z-drugs (aOR 2.07, 1.58 – 2.72), and antipsychotics (aOR 1.84, 1.48 – 2.28) were associated with fatal poisonings. Stratifying the outcome by intent, valproate was associated with a decreased risk of accidental fatal poisoning (aOR 0.50, 0.31 – 0.83) and carbamazepine was with an increased risk of fatal suicidal poisonings (aOR 2.20, 1.13 – 4.28).

Conclusions: Several medicines used by individuals with bipolar disorder were associated with fatal poisoning due to medicines or illicit substances. With the exception of carbamazepine, there was no evidence of an increased risk for other mood stabilizers. While the observed associations may reflect either greater toxicity in overdose or underlying clinical severity among users of these medicines, increased caution is warranted when prescribing them to individuals at elevated risk of suicidal or accidental overdose.

References

No references