B96-10 Randomized Trial of a Scalable, Interactive Digital Tool to Support Surrogate Decision-Makers of Critically Ill Adults at High Risk of Death or Severe Functional Impairment

Rationale

Prior family support interventions in ICUs have often required adding trained personnel to ICU teams, making them expensive and difficult to scale. We sought to determine whether a scalable intervention combining an interactive digital family support tool (FST) with proactively scheduled family meetings improves patient- and family-centeredness of care and other key outcomes for critically ill adults at high risk of death or severe functional impairment.

Methods

We conducted a multicenter, patient-randomized efficacy trial in 20 ICUs at 6 hospitals across the United States. Eligible patients were ≥18 years old, lacked decision-making capacity, and were at high risk (>40%) of in-hospital death or severe functional impairment. Patients were randomized 1:1 to the intervention versus usual care. The intervention consisted of: 1) unlimited surrogate access to the interactive, web-based FST throughout hospitalization, with prompted use upon enrollment and before family meetings; 2) proactively scheduled family meetings within 2 days of enrollment and at least weekly thereafter; and 3) prior to each family meeting, a summary report was shared with clinicians containing surrogates’ main questions, prognostic expectations, information about patient values, and surrogates’ current perceptions about appropriate treatment goals. Control patients received usual ICU care. The primary outcome was patient- and family-centeredness of care assessed by surrogates at 3 months using the modified Patient Perceived Patient-Centeredness of Care (PPPC) scale (range 1-4, lower scores indicate more patient-centered care). Other outcomes included patients’ ICU and hospital length of stay, and hospital mortality. We conducted a prespecified subgroup analysis among patients with high palliative care needs (NEST score ≥30).

Results

We enrolled 371 patients and 457 surrogates. Mean patient age was 67.0±12.9 years, 47.7% were female, and mean APACHE II score was 22.4±8.3. Intervention fidelity was high: 88.8% accessed the tool before first family meeting and 81.2% had family meetings per protocol. There was no significant difference in the primary outcome (mean PPPC score: intervention 1.6±0.6 vs. control 1.5±0.6, p = 0.36). There were no significant differences in ICU length of stay (11.7±13.0 vs. 12.3±12.6 days, p = 0.70), hospital length of stay (21.4±22.4 vs. 23.9±26.4 days, p = 0.58), or in-hospital mortality (37.2% vs. 38.8%, p = 0.82). Among patients with high palliative care needs, (NEST ≥30, n = 195) there was similarly no difference in patient-centeredness of care or length of stay.

Conclusions

A scalable intervention combining a digital family support tool with proactively scheduled family meetings did not improve patient- and family-centeredness of care or decrease hospital length of stay.

This abstract is funded by: NIA