PURPOSE: To develop an instrument for use at ICU discharge for prediction of psychological problems in ICU survivors.

METHODS: Multinational, prospective cohort study in ten general ICUs in secondary and tertiary care hospitals in Sweden, Denmark and the Netherlands. Adult patients with an ICU stay ≥ 12 h were eligible for inclusion. Patients in need of neurointensive care, with documented cognitive impairment, unable to communicate in the local language, without a home address or with more than one limitation of therapy were excluded. Primary outcome was psychological morbidity 3 months after ICU discharge, defined as Hospital Anxiety and Depression Scale (HADS) subscale score ≥ 11 or Post-traumatic Stress Symptoms Checklist-14 (PTSS-14) part B score > 45.

RESULTS: A total of 572 patients were included and 78% of patients alive at follow-up responded to questionnaires. Twenty percent were classified as having psychological problems post-ICU. Of 18 potential risk factors, four were included in the final prediction model after multivariable logistic regression analysis: symptoms of depression [odds ratio (OR) 1.29, 95% confidence interval (CI) 1.10-1.50], traumatic memories (OR 1.44, 95% CI 1.13-1.82), lack of social support (OR 3.28, 95% CI 1.47-7.32) and age (age-dependent OR, peak risk at age 49-65 years). The area under the receiver operating characteristics curve (AUC) for the instrument was 0.76 (95% CI 0.70-0.81).

CONCLUSIONS: We developed an instrument to predict individual patients' risk for psychological problems 3 months post-ICU, http://www.imm.ki.se/biostatistics/calculators/psychmorb/ . The instrument can be used for triage of patients for psychological ICU follow-up.

TRIAL REGISTRATION: The study was registered at clinicaltrials.gov, NCT02679157.