Medical end-of-life decision-making in a small resource-poor Caribbean country

BACKGROUND: In resource-poor countries, including in Latin American and the Caribbean, empirical information about the characteristics and incidence of medical end-of-life decisions (MELDs)-withholding or withdrawing potentially life-prolonging medical treatments-is largely absent.

METHODS: The aim was to describe the incidence and decision-making characteristics of MELDs taken prior to the death of people who died at home in Trinidad and Tobago (T&T). A mortality follow-back study was used where a representative sample of deaths occurring at home in 2018 was drawn from death certificates at the national death registry. The general practitioners who certified the deaths were sent a questionnaire.

RESULTS: The sample consisted of 309 adult deaths and the response rate was 31% (N=96). Physicians were: mostly male (79.2%), practiced medicine for more than twenty-years (63.5%), had no formal palliative care training (69.8%). Non-sudden deaths represented 76% (N=73), of these, medications to alleviate pain and symptoms in the last 7 days of life were administered in 65.8%, including opioids 21%. Potentially life-prolonging treatments were withheld in 9.6% but none withdrawn. No physician/patient discussions about various end-of-life treatment options occurred in 61.6%. Compared to physicians with no formal training in palliative care, those with training more often: prescribed or administered opioids in the last 7 days of life (35.7% vs. 11.1%, P=0.01), had discussions with patients about end-of-life treatment options (60.7% vs. 24.4%, P=0.002), and discussed medication use to alleviate pain and other symptoms with patients (50% vs. 17.8%, P=0.004).

CONCLUSIONS: Differences in the care and treatment general practitioners provided to their patients could be associated with them having been formally trained in palliative care. The necessary support to further develop palliative care in T&T is needed.