Tracheostomy has been relatively unpopular in Taiwan. Here, we investigate the impact of shared decision-making (SDM) for tracheostomy on critical patients undergoing prolonged intubation.

We retrospectively enrolled 1464 patients admitted to a tertiary medical centre intensive care unit (ICU) due to respiratory failure between April 2017 and April 2023. A 2-to-1 propensity-score with nearest-neighbour matching was used to balance covariates across SDM and non-SDM groups. Outcomes, including tracheostomy rate, intubation to tracheostomy time, mortality rate, ventilator weaning rate, ICU admission days, length of hospital stay, and 6-month post-discharge readmission rate were collected. Binary outcomes (Tracheostomy rate, in-hospital mortality, and 6-month readmission rate) were analysed using multivariable logistic regression, reported as odds ratios (ORs) with 95 % confidence intervals (CIs). Continuous outcomes (Intubation to Tr. T, ICU days, and length of stay) were assessed with generalized linear models, reported as regression coefficients (β) with 95 % CIs.

We found that SDM introduction was associated with a higher tracheostomy (44.3 % vs. 30.9 %, p < 0.001) and lower in-hospital mortality rates (15.7 % vs. 26.5 %, p = 0.004), but failed to demonstrate significant impacts on the intubation-to-tracheostomy time (25.8 ± 16.8 vs. 28.8 ± 17.6 d, p = 0.05) and the rest of the clinical outcomes. Subgroup analysis showed SDM had the greatest benefit to those with prior respiratory, neuromuscular, and malignant diseases.

Introducing SDM for prospective tracheostomy recipients is associated with better clinical outcomes in critical patients undergoing prolonged intubation.