Palliative care phenotypes among critically ill patients and family members: intensive care unit prospective cohort study

Abstract

Objective: Because the heterogeneity of patients in intensive care units (ICUs) and family members represents a challenge to palliative care delivery, we aimed to determine if distinct phenotypes of palliative care needs exist.

Methods: Prospective cohort study conducted among family members of adult patients undergoing mechanical ventilation in six medical and surgical ICUs. The primary outcome was palliative care need measured by the Needs at the End-of-Life Screening Tool (NEST, range from 0 (no need) to 130 (highest need)) completed 3 days after ICU admission. We also assessed quality of communication, clinician-family relationship and patient centredness of care. Latent class analysis of the NEST's 13 items was used to identify groups with similar patterns of serious palliative care needs.

Results: Among 257 family members, latent class analysis yielded a four-class model including complex communication needs (n=26, 10%; median NEST score 68.0), family spiritual and cultural needs (n=21, 8%; 40.0) and patient and family stress needs (n=43, 31%; 31.0), as well as a fourth group with fewer serious needs (n=167, 65%; 14.0). Interclass differences existed in quality of communication (median range 4.0-10.0, p<0.001), favourable clinician-family relationship (range 34.6%-98.2%, p<0.001) and both the patient centredness of care Eliciting Concerns (median range 4.0-5.0, p<0.001) and Decision-Making (median range 2.3-4.5, p<0.001) scales.

Conclusions: Four novel phenotypes of palliative care need were identified among ICU family members with distinct differences in the severity of needs and perceived quality of the clinician-family interaction. Knowledge of need class may help to inform the development of more person-centred models of ICU-based palliative care.

Keywords: clinical decisions; communication; family management.

Figure 1

Item response probabilities for latent classes of serious palliative care need. This figure shows item response probabilities for each of the four latent classes. This represents the probability that family members in a latent class report a serious level of need (item NEST score ≥5) for a particular item. For example, for the NEST item ‘I don’t understand clearly what to expect’, members of the decisional support and information needs class had a probability of 0.83 of reporting this as a serious need. NEST items that define each of the four latent classes are indicated by dark orange bars and bolded text, while NEST items that that do not define a latent class are indicated by yellow bars. ICU, intensive care unit; NEST, Needs at the End-of-Life Screening Tool.