Every year, many critically ill patients are admitted to an intensive care unit (ICU). Most stay only a few days, but a smaller group stays much longer. In Sweden, patients treated in the ICU for more than 10 days account for about 35 percent of all ICU days. Little is known about why some patients need prolonged intensive care, or how their lives are affected afterwards.
LIFE-ICU is a planned national observational cohort study of adult patients in general intensive care. This record covers the feasibility study that precedes the main study, including its pilot part.
The feasibility study does not change the care patients receive. About 200 adult patients who have been treated in a general ICU for more than four days will be included consecutively at the participating ICUs. For each patient, a designated next of kin, usually the family member or friend most involved during the ICU stay, is invited to take part. Information is collected from the medical record about health before admission and about the ICU stay, for example organ support, delirium, nutrition and mobilisation. At ICU discharge, patients answer four questions about distressing memories from the ICU.
Six months after final discharge from intensive care, patients are contacted by letter and then by telephone. They answer questions about health status, quality of life, anxiety and depression, post-traumatic stress, fatigue, life satisfaction, frailty, overall function and occupational status. Designated next of kin answer questions about their own life satisfaction, post-traumatic stress and caregiver burden, provided that the patient is alive at the time of follow-up. A selection of patients and next of kin are also invited to an interview about their experiences of intensive care and of recovery. Staff who collect the data answer an anonymous questionnaire about how the data collection worked.
The study compares patients whose total ICU stay is more than 10 days with patients whose stay is more than 4 days and at most 10 days.
The main purpose of the feasibility study is to find out whether the planned data collection works in practice. This is assessed against criteria set in advance: how many of the eligible patients can actually be included, how complete the collected data are, and how many patients and next of kin respond at six months. The pilot part also provides the estimates needed to plan the main study, such as the proportion of patients who stay in the ICU for more than 10 days and how the measured factors are distributed.
Hypothesis: the feasibility study tests the hypothesis that the planned data collection is feasible within routine intensive care and yields data of sufficient completeness and quality to inform the design of the main study. The main study is designed to test the hypothesis that factors present before ICU admission and during the first days of intensive care contribute to a prolonged ICU stay, that some of these factors can be modified, and that they are also associated with worse long-term outcome. The feasibility study is not powered to test this hypothesis.
The long-term goal is better knowledge about prolonged intensive care, so that prognostic information can become more accurate, care can be planned more individually, and patients and next of kin can receive better support during and after the ICU stay.
ICU stay of more than 4 days (96 hours) in a participating general intensive care unit (ICU)
Age 18 years or older
The first full 4 days of the current episode of intensive care were delivered at one or more participating ICUs. Patients transferred in from a non-study ICU during the first 4 days are not eligible.
No formal decision to withdraw life sustaining treatment taken before 4 days of ICU care
6 days (144 hours) or more since ICU admission at the time of inclusion
The patient in any form expresses unwillingness to take part
Age 18 years or older
Designated next of kin of an included patient, preferably the next of kin most involved in the patient's care in the ICU, or another person in line with the wishes of the patient and next of kin
Adult patients cared for in general intensive care units at participating university, county and district hospitals in Sweden, delivering mixed medical, surgical and trauma intensive care. All patients whose ICU stay exceeds 4 days are screened consecutively during the inclusion period, irrespective of admission diagnosis, so that the cohort reflects the case mix of routine Swedish general intensive care rather than a selected diagnostic group. Specialised units, that is cardiothoracic, neurointensive, burn and paediatric intensive care units, do not include patients. Designated next of kin of included patients take part in their own right, preferably the next of kin most involved in the patient's care in the ICU.
Cohort
Prospective
Not listed
Inclusion rate: proportion of eligible screened patients who are included
Denominator is all patients screened during the inclusion period who fulfil the eligibility criteria other than consent and no objection, that is age, timing, ICU episode and treatment limitation. Numerator is those included. Screening is documented in a log at each site with one row per screened patient. Reasons for non-inclusion are categorised as consent requirements not met, patient objects, next of kin objects, decision to withdraw life sustaining treatment taken before eligibility, logistic reasons, and other. The proportion is reported overall and per site with a 95 percent confidence interval. Progression criteria: at least 80 percent, proceed without change. 60 to 79 percent, proceed after revision of screening and consent routines. Below 60 percent, redesign before the main study.
Data completeness: proportion of applicable required fields completed in the electronic case report form
Denominator is all required fields, and conditional fields rendered applicable by branching logic, for included patients across the baseline form, the retrospective forms for ICU days 0 to 3, the daily and per shift forms from day 4 until ICU discharge, and the discharge form. Numerator is fields containing a valid value. Completeness is reported overall, per form, per data domain and by inclusion route with a 95 percent confidence interval, together with patterns of missing data and the proportion of patients with a complete record. Progression criteria: at least 95 percent, proceed without change. 85 to 94 percent, proceed after revision of the form and the data collection manual. Below 85 percent, redesign before the main study.
Six month follow-up response rate in patients and designated next of kin
Numerator is participants who complete or return at least one follow-up instrument. Denominator for patients is all included patients alive at follow-up, including those unable to complete the questionnaires in Swedish. Denominator for designated next of kin is all identified designated next of kin who have consented and whose patient is alive at follow-up. Contact follows the predefined letter, telephone and reminder routine for each group. Rates are reported separately with 95 percent confidence intervals, together with the proportion completing all instruments, the number of contact attempts, reasons for non-response, and the number of designated next of kin who declined consent at follow-up. Progression criteria for patients: at least 75 percent, proceed. 60 to 74 percent, revise routines. Below 60 percent, redesign. Levels for designated next of kin are 10 percentage points lower.
Timeliness of data entry into the electronic case report form
Days from generation of a data item at the bedside to entry of that item in REDCap, reported as median and interquartile range overall, per form and per site. Timeliness is interpreted together with data completeness when the data collection routines of the main study are specified.
Proportion of patients with a total ICU length of stay of more than 10 days
Proportion of included patients whose total ICU length of stay is more than 10 days (240 hours), reported with a 95 percent confidence interval. Total ICU length of stay comprises all consecutive ICU care within the same continuous episode of critical illness, including care at non-study ICUs after transfer. This proportion is the main input for the sample size calculation of the main study. Distributions of candidate predictors and confounders, their variances and correlations, and exploratory regression estimates are reported separately as described in the detailed description.
Record linkage to the Swedish Intensive Care Registry
Proportion of included patients for whom a corresponding record can be retrieved from the Swedish Intensive Care Registry using the secondary key, reported with a 95 percent confidence interval. Reasons for failed linkage are categorised. The proportion of patients whose total ICU length of stay could only be established through registry data, that is patients transferred to a non-study ICU, is also reported.
Distribution of inclusion routes
Proportion of included patients in each inclusion route, that is written informed consent from the patient at inclusion, consultation with the next of kin, or inclusion without prior consultation because no next of kin or legal representative could be identified despite documented attempts. Reported with 95 percent confidence intervals, overall and per site. Patients included without prior consultation are compared with the remaining cohort as described in the detailed description.
Click on the lead sponsor to view all of their trials.
Uppsala University
Lead sponsor
Region Gävleborg
Collaborator
Dalarna County Council, Sweden
Collaborator
Region Uppsala Public Health and Care
Collaborator
This trial is not recruiting at the moment. You can still explore other options:
