Introduction: Severity of illness, measured by the Acute Physiology and Chronic Health Evaluation (APACHE) score, is expected to correlate with length of mechanical ventilation (LOV). Since increasing age contributes to higher APACHE scores, older patients would be expected to have longer LOV. We conducted a retrospective study to examine the relationship between APACHE score, age, and LOV. We also analyzed the correlation between LOV and age-adjusted APACHE scores, calculated by subtracting the age-derived points from each patient’s APACHE score, to assess physiologic illness severity independent of age. Methods: We analyzed adult ICU patients who received mechanical ventilation across hospitals in our network between October 2015 and December 2024. Patients were divided into five age groups: 16–44, 45–59, 60–69, 70–79, and ≥80 years. Median LOV across these age groups was compared using the Kruskal-Wallis test. Both unadjusted and age-adjusted APACHE scores were calculated for all groups. Results: A total of 15,292 patients were included: 1,958 (16–44), 2,533 (45–59), 3,590 (60–69), 3,688 (70–79), and 3,523 (≥80). Median LOV for each group was 1.57, 1.32, 1.21, 1.29, and 1.46 days, respectively. Mean APACHE scores were 60, 64, 68, 74, and 80, while age-adjusted mean APACHE scores were 60, 59, 56, 57, and 59. Despite having comparable age-adjusted APACHE scores, suggesting similar physiological illness severity, the youngest group had a significantly longer LOV when compared independently with each of the older age groups (all p values < 0.05). Conclusions: Younger adults have longer LOV compared to older patients with similar illness severity. This may reflect age-related differences in provider extubation practices, with clinicians potentially being more cautious in extubating younger patients and/or more aggressive in extubating older patients. Differences in admission diagnoses between the age groups, not fully captured by the APACHE score, may also contribute. Finally, higher mortality in older patients, if present, may have artificially shortened LOV in that population, potentially skewing the data. Further analysis of this data using ventilator-free days will be important to confirm whether these findings persist after adjusting for mortality.
SYNN et al. (Sun,) studied this question.